This study contributes to the literature of migration studies by addressing the question: why does international migration persist despite welfare improvements in migrant-sending countries? We propose that the human rights condition of the origin countries is an important determinant of global migration. Although the human rights issue is not new to researchers in migration studies, the concern is primarily about the rights of migrants, refugees, asylum seekers or migrant workers in a host country. We undertake a bilateral panel data analysis to examine the pattern of global bilateral migration between 1995 and 2010. We find that international migration is positively associated with human rights conditions and income. Similar results are also obtained when we control for multilateral resistance and possible sample selection biases in a panel context. Our study implies that efforts to promote human rights may also be assessed in relation to their contribution to migration flows.

Introduction

Scholarly studies consistently show that socio-economic and political factors are major motivations for international migration (Baudassé et al., 2018). While the importance of economic well-being is recognised by scholars across academic disciplines, the relevance of political factors, except violent conflicts, are more controversial. For example, studies by Jones (1989) and Radnitz (2006) found that political factors were not sufficient to explain migration in El Salvador and post-Soviet Uzbekistan. Jones (1989) argued that it was economic setbacks that triggered Salvadoran migration to the United States. Similarly, Radnitz (2006) considered that political factors were rarely sufficient to produce migration. Socio-political factors like nationalism influenced migration decisions only if they affected people's material well-being. On the other hand, a recent review article by Baudassé et al. (2018) shows that political factors play significant roles in explaining international migration.

Despite the existence of evidence for the importance of a wide array of socio-political factors, previous studies mainly focus only on a number of them: democracy (Narayan and Smyth, 2005), political and civil rights (Karemera et al., 2000), violent conflicts (Ibáñez and Vélez, 2008) and government effectiveness (Ariu et al., 2016). This study examines a different political factor in determining a migrant's decision to move out: human rights. While the human rights issue is not new to researchers in migration studies, the concern is primarily about the rights of migrants, refugees, asylum seekers and migrant workers in a destination country. For example, Gallagher (2002) addressed the issue of human trafficking; Crush (2001) discussed intolerance and discrimination against migrants in South Africa; Cholewinski (1997) assessed various rights of migrant workers in their country of employment. However, respect for human rights is seldom treated as a potential driver of emigration (Morrison and May, 1994; Bilodeau, 2008). Although the issue of human rights is extensively discussed in Chapter 5 of the 2009 Human Development Report, Overcoming Barriers: Human Mobility and Development, the human rights issue is not recognized as a determinant of migration and was never mentioned in the discussion on motivations for migration in Chapter 2 of the report.

Only a handful of studies explicitly consider human rights as a motivation for international migration and focus exclusively on the effects of “repression” or “physical integrity” on migration (Schmeidl, 1997; Moore and Shellman, 2004; Bilodeau, 2008). This study extends the understanding of the concept of human rights beyond the “physical integrity” definition and expands the scope to include the rights to social, economic and political freedom. Most people are unlikely to be the target of the state or face state persecution. That being said, individuals are distressed if their rights are limited and value the freedom to exercise their liberty to maximize their potential with minimum constraint. Accordingly, we consider the aspiration of individuals to improve their living and to thrive, complementing economic motives, an important determinant of migration (Carling, 2002; De Haas, 2010). Therefore, although we agree that fear of persecution is a driver of international migration, this article adopts a broader approach. As human rights are also about human development, we use the latent scores estimated by Fariss (2014) as an indicator of human rights practices in a country, because the indicator captures an additional, important aspect of human rights: empowerment. An objective of the rights-based approach to development is to allow marginalized people to take control of their lives and to claim their entitlements (Nankani et al., 2005). In this way, this study departs from the “physical integrity” approach and is more general. This study specifically looks at the effect of respect for human rights on migration and defines respect for human rights as human rights practices in a country, which encompass both repression and empowerment. Contrary to human rights protection, which is mainly about measures (e.g. law) that aim to protect the rights of a person (e.g. Hathaway, 1991), respect for human rights concentrates on actual practices and behaviours instead of intention. We take this approach because it is difficult to assess enforcement and compliance of human rights law comprehensively in a macro panel analysis.

Apart from the ways of conceptualizing human rights, most existing studies focus solely on forced migration (e.g. Morrison and May, 1994; Abusharaf, 1997; Schmeidl, 1997; Moore and Shellman, 2004; Bilodeau, 2008). For instance, Moore and Shellman (2004) looked into the impact of political persecution on forced migration and argued that political violence alters people's expectations regarding their chances of being prosecuted. The fear of falling victim to state violence drives people to take refuge in another country. As we consider aspiration and human development important motivations for migration, this study goes beyond asylum seeking and considers regular migration. Therefore, this article argues that the effect of human rights is more general than previous studies suggested.

Our empirical analysis shows that, in relation to international migration, respect for human rights is as important as other well-known economic determinants. One may wonder to what extent is the human rights factor independent of the economic factors (e.g. Jones, 1989; Radnitz, 2006). We agree that economic prospects and the amount of resources available to a potential migrant are crucially important to turn intention into action, but due to the fear of prosecution (Moore and Shellman, 2004) and the desire to have greater freedom in their lives, economic concerns do not eliminate the role of human rights in the decision-making process of a potential migrant. As we will elaborate in the next section, many people from emerging economies emigrate because of the greater political freedom in their new homes (Zweig, 2006). The human rights factor is not only a sufficient condition of migration. It could also be the reason behind poor economic well-being. For instance, poor economic welfare could be a result of political inequality and social exclusion. This explains why individuals belonging to certain ethnic minorities choose to leave their homes and migrate to other countries.

Our finding leads to an important policy implication. The government of a net migrant-receiving country may factor the secondary effects of human rights conditions on emigration in their foreign policies regarding human rights issues. Although the effectiveness of external pressure on human rights conditions remains questionable (Portela, 2010), and there is only limited evidence for it, examples from the US and the European Union show that external actors could sensitise human rights conditions of a country in a meaningful way. “Naming and shaming” and conditioning, which will be further particularized in the discussion section, are two possible approaches to achieve this goal. In this light, this study provides an additional reason why a country may want the human rights conditions of another country to improve. It is because human rights abuses can induce emigration and increase the social, political and welfare costs at the destination countries.

The rest of this study is structured as follows. After elaborating on the hypothesized relationship between human rights and migration in the next section, we will then describe the research design that tests our argument and present the results. The policy implication will be further discussed in the discussion section before the conclusion.

Migration and Human Rights

Existing studies generally agree that regular migrants choose destinations that maximise their economic opportunities. The search for a greater employment prospect and higher earnings are two of the most important determinants of migration (e.g. Harris and Todaro, 1970; Greenwood et al., 1986; McKenzie et al., 2014). Recent contributions have considered a wider set of factors. For example, some sociologists emphasised the social dimension of migration. They considered cultural, communal and family ties vital factors in perpetuating migration between countries (Boyd, 1989). Based on the network theory, Palloni et al. (2001) argued that friends and relatives provide information and support to the newly migrated. This enables newcomers to gain access to jobs, an arrangement that helps to reduce the risks and costs of migration (Massey, 1990). Some geographers examined the role of climate change and observed that exposure to different natural hazards, such as drought and flooding is associated with human migration (Hugo, 1996; McLeman and Smit, 2006). Some political scientists argued that migration could be induced by violent conflicts. Migrants flee from their home in order to save their lives (Ibáñez and Vélez, 2008; Hatton, 2009; Bohra-Mishra and Massey, 2011). All in all, migration can be considered as an economic investment, a product of network behaviour and a survival strategy.

A country that has a large migrant outflow is not only characterized by low income, high unemployment rate, low quality of life and political instability, but also by a poor human rights record. Human rights can be broadly defined as a set of rights based on the principles of liberty and equality — which encompass various dimensions of human lives (Charvet and Kaczynska-Nay, 2008). The nature of the recent emigration from China illustrates this point particularly well. Although the Chinese economy has been on the rise since the early 1980s and performed particularly well in the 2000s, Chinese emigration also increased sharply during the same period. According to the World Bank (2018), China has recorded a net migration outflow since 1965 and sends hundred thousands of students overseas every year. Between 2000 and 2013, the country cumulatively sent 2.7 million students to study abroad. But during the same period, there were only about 1.3 million students decided to return to China (NBSC, 2015). The reasons for the brain drain are manifold. According to a survey on Chinese overseas students by Zweig (2006), the lack of political stability and political freedom were the top two reasons why students chose not to return and decided to settle in another country. Furthermore, both reasons were found to be more important than the factors of no opportunity for career advancement in China (ranked 3rd), poor working environment (4th) and low living standard (5th) (Zweig, 2006).

While respect for human rights is a strong factor, institutional violence and state persecution are decisive factors that drive people away from their homes. As Schmeidl (1997) showed, every year, the concerns about physical security drive thousands of refugees fleeing away from their homeland. Moore and Shellman (2004) argued that state repression is a visible public signal to a wide population. People who observe the signal will revise their beliefs about the threat to their physical integrity. When the subjective probability of being a victim of persecution becomes too high, people migrate (Moore and Shellman, 2004).

The argument can be generalized to cases where government is not the only relevant actor in human rights violations. For instance, extrajudicial killings are serious human rights violations, seen in Albania, Kosovo and Pakistan (United Nations General Assembly, 2013). “Honour killing” is regarded as a form of gender-based violence in Pakistan and some other Islamic countries. Individuals, especially females, who refuse arranged marriages, are under the risks of being pursued and killed by their family members, who consider themselves authorized to kill them in order to restore the honour of their families. As a result, many people leave their home countries and seek asylum in culturally more liberal countries (Plant, 2005). This type of non-state-led violence highlights how the respect for human rights in a country as a whole can influence one's propensity to emigrate.

In brief, this study considers human rights as the right to life, liberty and security. Accordingly, human rights abuses are not limited to acts that violate physical and personal integrity rights, exemplified by extrajudicial killing, torture and political persecution. It also includes restrictions on economic and political freedom. People do care about their rights being limited. They value the freedom to exercise their liberty to maximise their potential. As Carling (2002) pointed out, many migrants aspire to improve their living and to thrive. In this way, the inclusion of human rights considerations to the aspiration approach not only breaks down the economic/political and forced/voluntary migration dichotomies, it also offers a more complete understanding of the decision to migration.

Empirical Approach

Our econometric model, similar to the one in Ortega and Peri (2013), is a gravity-like bilateral specification between the origin country i and the destination country j:

$urn:x-wiley:00207985:media:imig12558:imig12558-math-0001$

where t indexes the year of observation, m_ijt is the size of the migration flow from origin i to destination j, d_ij is the geographical distance between the two locations, y_it, y_jt and Rights_it are, respectively, the income per capita and human rights scores of a country. The dummy variables language and colony take the value of one if i and j, respectively, share a common language and have past colonial relations. Finally, e_ijt is an error term.

There are three major empirical challenges associated with our analysis: multilateral resistances, zero migration flows and missing observations. We will discuss each of these issues and our empirical strategies below. As each strategy has its own limitation, our purpose is to show that our key findings are robust across different empirical methods used to cope with the above empirical challenges.

Multilateral resistance

Bilateral flows between two given countries are outcomes of different pull and push forces exhibited, not only by the two given locations but also by all the other countries in the world. For instance, within the context of international trade, Beckerman (1956) has pointed out the importance of the relative distance between two economies as opposed to absolute distance. This idea was later formalized in the gravity model of trade as “multilateral resistance” by Anderson and van Wincoop (2003). Multilateral resistance to migration was first addressed both theoretically and empirically by Bertoli and Moraga (2013). For bilateral migration flows between two locations, a multilateral resistance term (MRT) represents “the influence exerted by the opportunities (and barriers) to migrate to other destinations” (Bertoli and Moraga, 2013: 82). As a result, the omission of multilateral resistance in the estimation of a gravity model of migration will yield biased estimates (Beine et al., 2016).

Several empirical approaches have been proposed to control for multilateral resistances in gravity models. For example, Bertoli and Moraga (2013) proposed the use of the common correlated effects estimator by Pesaran (2006) for a long panel. Since we only have a small T in our data set, we take the approach suggested by Baier and Bergstrand (2009) and transform the distance variable using the weighted distances of the origin and destination locations to all other locations. Using migration flows as weights, the approach corresponds to redefining a distance variable d_ij for origin i and destination j as:

$urn:x-wiley:00207985:media:imig12558:imig12558-math-0002$

where $urn:x-wiley:00207985:media:imig12558:imig12558-math-0003$ is a weight, for n destination countries and m origin countries, and d_ij is the geographical distance of the origin country i to the destination country j.

Zero migration flows

In some cases, there is no migration recorded between two countries; for example, from Fiji to Angola or from Georgia to Bolivia. Because of the skewed distribution of the migration variable, researchers often choose to log-normalize the variable. But in the presence of migration flows of zero, a log transformation does not work because the logarithm of zero is mathematically undefined. In this study, we try different transformation methods and models in our estimation. Our first attempt, or a baseline, is not to log-normalize the variable and estimate the model with OLS. Second, we try the inverse hyperbolic sine (IHS) transformation, which is used to transform data on wealth (e.g. Carroll et al., 2003). More specifically, IHS transformation is defined as $urn:x-wiley:00207985:media:imig12558:imig12558-math-0004$ . Unlike a log transformation, the inverse hyperbolic sine transformation is defined when y is equal to zero. The estimated coefficient also has the same interpretation as the logarithmic counterpart (Pence, 2006). Third, we recode negative net migration flow (i.e. return migration) as zero and estimate the models with the Poisson pseudo-maximum likelihood (PPML) method, which is a standard approach in estimating trade gravity model (Santos Silva and Tenreyro, 2006). Differed from the OLS and IHS approaches, this strategy treats return migration as a phenomenon distinct from regular migration and does not mix them in one model.

Missing data

Missing data are a major empirical challenge in research using a gravity model. We employ two different strategies to ensure our findings robust to the missing data problem. Our first approach is to replace them with zeros, a practice sometimes adopted in the literature (Santos Silva and Tenreyro, 2006). It is unlikely that there is no migration at all in those missing cells. Therefore, replacing them with zeros is likely to under-estimate the true effect of our key variable, and is against our hypothesis. For this reason, the strategy can be a useful robustness check to the key findings.

Our second approach is to treat the model in the context of a sample selection problem. Sample selection bias may arise when there are missing data in migration flows between two countries. Therefore, the probability of migrating would be an omitted variable in the model. In fact, data of immigrants from most developing countries are missing. To correct for the selection bias, we use Heckman's (1979) sample selection method. However, as Wooldridge (1995) noted, in a panel-data setting, simply adding the estimated inverse Mills ratio to the main equation and using fixed-effects estimator will produce inconsistent estimates. To tackle this empirical challenge, we employ the solution proposed by Wooldridge (1995) to correct for sample selection in a fixed-effects context. More specifically, we first model selection with a probit model using all independent variables in our core model, as well as the freedom of domestic and foreign movement dummies as additional selection variables. Then we obtain the inverse Mills ratio, $urn:x-wiley:00207985:media:imig12558:imig12558-math-0005$ , and run the following pooled OLS regression using the selected sample:

$urn:x-wiley:00207985:media:imig12558:imig12558-math-0006$

where X is the vector of independent variables and d2_t through dT_t are time dummies. The standard errors from the OLS, however, are incorrect. Therefore, we use panel bootstrapping suggested by Semykina and Wooldridge (2010), with 1,000 repetitions, to obtain the correct standard errors.

Data

The data used in this study come from various sources. The dependent variable, the flow of migrants from the country of origin i to the destination country j is based on the United Nations Population Division (2015), which has been used to code other data set such as Özden et al. (2011). Due to the subject of interest, we examined only migration from developing countries. On the other hand, we allow all developing countries in the data set as potential destinations. The unbalanced panel includes 179 destination countries and 156 countries of origin, spanning from 1995 to 2010, over a 5-year interval (i.e. T = 4). The panel data analysis ends in 2010 because the key independent variable, Rights_it, ends in 2014.

The main explanatory variable, the human rights score (Rights_it), comes from Fariss (2014). It is a latent score generated based on data from various sources which include, for example, reports from the US State Department, Amnesty International, the Political Terror Scale Project (Gibney and Dalton, 1996; Wood and Gibney, 2010), and the CIRI Human Rights data set (Cingranelli and Richards, 2010). A higher score indicates that a country has better respect for human rights and is politically less repressive than those with a lower score. Existing studies mainly rely on few indicators in their studies; for instance, Moore and Shellman (2004) use the Political Terror Scale in their study. In contrast, the indicator used in this study embraces a multidimensional approach and contains information about freedom of association, freedom of speech, rights of workers, etc. The analysis is also extended to the other general indicators prepared by the US State Department and Amnesty International, and specific indicators in the CIRI Human Rights data set, the Political Terror Scale and the Freedom House indices.

The other key independent variables, y_it and y_jt, are GDP per capita (constant 2011 international dollars, in natural logarithm) of the destination and origin countries. As economic theory predicts (Mayda, 2010), migrants respond to economic opportunities available abroad and move from a low-income region to a high-income region. Data come from the World Bank (2017).

We also include a number of control variables, described below, that are commonly included in an analysis of migration. Table 1 provides the summary statistics of the variables.

Table 1. Descriptive Statistics

Variable	N	Mean	Std. Dev.	Min	Max
Flow (in thousands)	28,524	1.96	34.25	−1014.57	2574.69
Flow (IHS transformed)	28,524	0.30	1.16	−7.62	8.55
Flow (truncated at 0)	28,524	2.86	30.90	0	2574.69
ln y_i	28,524	8.70	1.12	5.51	11.74
ln y_j	28,524	9.70	1.02	5.51	11.74
Rights_it (z-score)	28,524	−0.30	0.83	−2.31	2.44
Language_ij	28,524	0.20	0.40	0	1
ln d_ij	28,524	8.35	0.97	4.09	9.89
ln d_ij (MRT)	28,524	8.17	0.96	4.04	9.84
Colony_ij	28,524	0.03	0.17	0	1
Conflict_it	28,524	0.07	0.26	0	1
Foreign Movement_it	28,168	1.37	0.74	0	2
Domestic Movement_it	28,168	1.35	0.77	0	2
Rights_it	28,524	-0.30	0.83	−2.31	2.44
Amnesty_it	24,663	3.39	1.12	1	5
State Department_it	27,745	3.61	1.17	1	5
Civil liberties_it	28,524	4.14	1.69	1	7
Political rights_it	28,524	4.15	2.08	1	7
Press freedom_it	28,326	−0.21	0.90	−2.18	1.66
Physical integrity_it	26,946	4.11	2.24	0	8
Disappearence_it	26,946	1.56	0.71	0	2
Extrajudicial killing_it	26,946	1.09	0.77	0	2
Political imprisonment_it	26,946	0.97	0.82	0	2
Torture_it	26,946	0.49	0.61	0	2
Freedom of Assembly_it	26,946	1.09	0.79	0	2
Freedom of Speech_it	26,946	0.85	0.69	0	2
Electoral Determination_it	26,946	1.06	0.80	0	2
Freedom of Religion_it	28,158	1.12	0.87	0	2
Worker's Rights_it	26,946	0.78	0.66	0	2
Judiciary Independence_it	28,168	0.88	0.78	0	2

Distance (d_ijt): distance between the capitals of the origin country and the destination country. Data come from Head et al. (2010). The geographical distance between two countries has been identified as one of the most important determinants of international migration (Mayda, 2010; Beine et al., 2016). It is positively related to the costs of migration and hence is regarded as a major barrier to migration. The variable is expected to have a negative impact on the dependent variable.
Colonial ties (Colony_ij): a dummy variable coded as one if the country dyad has had ever in a colonial relationship. Data come from Head et al. (2010). Emigrants from the previous colonies, shall they choose to migrate to an advanced economy, often view coloniser countries as attractive destinations to migrate. This may be due to the commonality in terms of culture and political institutions relative to an advanced economy which has not in a colonial relationship. The variable is expected to have a positive impact on the dependent variable.
Common language (Language_ij): a dummy variable equal to one if the countries in pair use the same languages spoken by over 9 per cent of the population in both countries. Data come from Head et al. (2010). Stronger language skills provide migrants a competitive advantage in job seeking (Dustmann and Fabbri, 2003). Since migration is also a social phenomenon, strong language skills facilitate integration process by improving their social experiences.
Freedom of domestic and foreign movement: indices which denote the levels of freedom to move within and leave or enter the country of origin. Data come from the CIRI Human Rights data set (Cingranelli and Richards, 2010). We use these indicators as the selection variables in estimating the Heckman models. Countries that restrict their citizens to move abroad are expected to have low or no international migration.

Empirical Results

The pooled OLS result based on the core model is presented in the first column of Table 2. The basic model yields the expected results. The better the human rights situation of a country is, the fewer people would choose to emigrate. The estimate is statistically significant at 0.1 per cent level. In terms of economic significance, since the Rights indicator is a latent variable, the interpretation of the coefficient is not intuitive. So we standardized the variable by subtracting the mean and dividing by the standard deviation of the variable. When the Rights indicator increases by one standard deviation, emigration flow decreases by an average of about 2,200 people, which is slightly larger than the mean of 1,960 people. Note that the number is in the unit of country dyad in a 5-year internal. With about 200 potential countries of origin, the actual impact is about 2,200 × 200 = 440,000 migrants. Income level at the destination country and the geographical distance between two countries are also good predictors of emigration flow. We extend our model by taking the multilateral resistance terms into account. The results are reported in column 2, which are similar to those in column 1. The results regarding per capita income and individual freedoms are consistent with our earlier observations as well.

Table 2. Effects of Respect for Human Rights on Migration

	(1)	(2)	(3)	(4)
	OLS	OLS	j-FE (LSDV)	Selection (Wooldridge, 1995)
Rights _it	−2.190*** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	−2.225*** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	−1.933** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	−1.719** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.
Rights _it	(0.621)	(0.628)	(0.649)	(0.660)
ln y _it	0.287	0.213	0.467	−0.315
ln y _it	(0.321)	(0.308)	(0.343)	(0.304)
ln y _jt	2.761*** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	3.022*** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	−0.895	4.334*** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.
ln y _jt	(0.467)	(0.532)	(0.741)	(0.747)
Language _ij	4.378** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	3.305* Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	1.820	2.103
Language _ij	(1.672)	(1.480)	(1.485)	(1.377)
ln dist _ijt	−0.700* Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.
ln dist _ijt	(0.353)
ln dist_ijt (MRT)		−2.612*** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	−1.779** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	−1.667* Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.
ln dist_ijt (MRT)		(0.689)	(0.576)	(0.662)
Colony _ij	−0.102	−0.222	0.540	(omitted)# Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.
Colony _ij	(3.733)	(3.729)	(4.112)
Conflict _it	−2.051	−2.128	−2.613	−1.273
Conflict _it	(1.377)	(1.394)	(1.416)	(1.457)
Constant	−23.31*** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	−9.455	15.91* Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.	−21.41*** Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: p < 0.05, p < 0.01, *p < 0.001. #The variable Colony*_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.
Constant	(6.803)	(4.915)	(7.678)	(6.220)
Observations	28,524	28,524	28,524	124,397
Number of pairs	7,306	7,306	7,306	27,769
Year effects	Yes	Yes	Yes	Yes
Destination-specific year effects	-	-	Yes	-
SE clustered by	Pair	Pair	Pair	-

Note:

Robust clustered standard errors in parentheses. Bootstrapped standard errors in parentheses in the selection model. Significance level: *p < 0.05, **p < 0.01, ***p < 0.001. #The variable Colony_ij is omitted in the selection model because it predicts (non)selection in the first-stage probit perfectly.

One major data limitation is that we do not have data on the immigration policy of each country dyad during the sampling periods. For instance, we do not know the evolution of immigration policy of Argentina to China or of South Africa to Mali between 1995 and 2010. In fact, existing quantitative studies only manage to code the variable for 15 Organization for Economic Co-operation and Development (OECD) countries (Ortega and Peri, 2013). To mitigate the associated bias, we interact the destination country dummies with year to introduce the year-specific destination fixed-effects (i.e. the destination fixed-effects can vary across years). The results are presented in column 3. Most importantly, poor respect for human rights has a negative and significant impact on the migration from i to j. In other words, the greater the respect for human rights in the origin economy, the smaller is the migration flow between two countries after controlling for destination-specific time effect. Model controlling for dyadic fixed-effects gives similar results (not shown in Table 2).

Another extension of our model introduced a sample selection component. Using the sample selection model (Heckman, 1979) and extending it to the panel setting (Wooldridge, 1995), we internalize the information conveyed in bilateral migration flows with missing observations. Results from the three MRT variants of the Heckman model are presented in column (4) in Table 2. Although the size of the estimate is smaller (i.e. equal to 1.72, which is slightly below the mean of 1.96), the results from a Heckman model reinforce our previous findings and add further information about the role of the model variables in relation to the probability of migration. Per capita income and individual freedom not only increase the magnitude of emigration, but also increase the probability of migration to take place between i and j. This finding is robust to the Baier and Bergstrand (2009) approach to control for multilateral resistance. In terms of the substantive effect, if respect for human rights in the origin countries improves, emigration will decrease by more than 1,200 people per country dyad. The results from the Heckman model also suggest that sample selection is indeed an issue that could lead to bias if not controlled for, as the inverse Mills ratios are significant for the models (not shown in Table 2).

The dependent variable is featured by a point mass at zero. As a robustness check, we extend our analysis by rescaling the variable using the inverse hyperbolic sine (IHS) transformation and estimate the new model with OLS (column 5) and with year-specific destination fixed-effects using the least-square dummy variable (LSDV) approach (column 6). The results of the checks are reported in Table 3. We also applied a new estimation method, the Poisson pseudo-maximum likelihood (PPML), to estimate the effects of income and respect for human rights on migration. Consistent with our hypothesis, respect for human rights has the expected negative effect on migration flow.

Table 3. Estimation Results Using Transformed Migration Flow Variables

	(5)	(6)	(7)	(8)
	OLS (IHS)	j-FE (LSDV) (IHS)	PPML (Flow)	OLS (Flow)
Rights _it	−0.199*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	−0.196*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	−1.068*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	−0.402*** Significance level: p < 0.05, p < 0.01, **p < 0.001.
Rights _it	(0.014)	(0.012)	(0.139)	(0.110)
ln y _it	−0.0140	0.029*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	−0.193* Significance level: p < 0.05, p < 0.01, **p < 0.001.	0.102
ln y _it	(0.009)	(0.008)	(0.097)	(0.071)
ln y _jt	0.275*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	−0.127	0.728*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	0.512*** Significance level: p < 0.05, p < 0.01, **p < 0.001.
ln y _jt	(0.013)	(0.116)	(0.076)	(0.079)
Language _ij	0.265*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	0.123** Significance level: p < 0.05, p < 0.01, **p < 0.001.	0.511* Significance level: p < 0.05, p < 0.01, **p < 0.001.	0.921* Significance level: p < 0.05, p < 0.01, **p < 0.001.
Language _ij	(0.036)	(0.038)	(0.238)	(0.399)
ln dist_ijt (MRT)	−0.193*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	−0.157*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	−1.200*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	−0.893*** Significance level: p < 0.05, p < 0.01, **p < 0.001.
ln dist_ijt (MRT)	(0.016)	(0.016)	(0.071)	(0.241)
Colony _ij	0.214	0.138	1.228*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	1.277
Colony _ij	(0.138)	(0.135)	(0.269)	(2.767)
Conflict _it	−0.0834	−0.098* Significance level: p < 0.05, p < 0.01, **p < 0.001.	−0.419	−0.433
Conflict _it	(0.045)	(0.039)	(0.272)	(0.378)
Constant	−0.820*** Significance level: p < 0.05, p < 0.01, **p < 0.001.	1.955* Significance level: p < 0.05, p < 0.01, **p < 0.001.	4.209** Significance level: p < 0.05, p < 0.01, **p < 0.001.	2.125
Constant	(0.201)	(0.898)	(1.577)	(1.527)
Observations	28,524	28,524	28,524	126,554
Number of pairs	7,306	7,306	7,306	27,769
Year effects	Yes	Yes	Yes	Yes
Destination-specific year effects	–	Yes	–	–
SE clustered by	Pair	Pair	Pair	Pair

Note:

Dependent variable in models (5) and (6): inverse hyperbolic sine (IHS) transformed migration flow. Dependent variable in model (7): truncated migration flow. Dependent variable in model (8): missing value replaced with 0. Robust clustered standard errors in parentheses.
Significance level: *p < 0.05, **p < 0.01, ***p < 0.001.

Finally, we use different human rights indicators to tease out different dimensions of human rights. Estimations are based on the selection model. Due to limited space, we report only the estimated coefficients of different indicators, with their 95 per cent confidence intervals in Figures 1 (different dimensions of human rights) and 2 (different general indicators). Except for the indicators of physical integrity, freedom in media and general human rights score, indicators are in a scale of 0-3 (CIRI: all indicators shown in Figure 1), 1-7 (Freedom House: general indicators on civil liberties and political rights in Figure 2) and 1-5 (Political Terror Scale: general indicators from Amnesty International and State Department in Figure 2). For the ease of interpretation, we rescaled some of the variables. Therefore, a higher score always means better respect and the lowest possible point is the baseline. At various degrees, all indicators have statistically significant effects on migration flow. In comparison, rights of personal security, such as freedom from extrajudicial killing, have the greatest effects on emigration. Though significant, economic rights (e.g. worker's rights), social rights (e.g. freedom of speech and media) and political rights (e.g. judiciary independence and rights to self-determination) have smaller though statistically significant effects on migration outflow.

**Figure 1**
Open in figure viewer PowerPoint

Effects of Human Rights on Migration, Different Dimensions

**Figure 2**
Open in figure viewer PowerPoint

Effects of Human Rights on Migration, Different General Indicators

Discussion

Many studies have shown that migrants probably left their birth places for better economic prospects. Ignoring the potential effect of human rights, however, may confine the set of solutions to the challenge of massive migration inflow to a country. The findings of this article suggest that there may be additional ways of governing the migration challenge; that is, promoting human rights of a developing country.

The effectiveness of external pressure on human rights conditions remains a subject of debate (e.g. Portela, 2010). Some suggested that, with proper designs and supporting monitoring mechanisms, external actors may use social market power as leverage to promote human rights (Kim and Sikkink, 2010). But there is only limited evidence for the external approach. For example, Keith (1999) found that becoming a member of an international covenant has no observable impact on state behaviours. Neumayer (2005) showed that domestic conditions are a more decisive factor than the international regime in determining the human rights conditions of a country (see also Cole, 2015, for a brief review on the ineffectiveness of the external approach). For the purpose of discussion, we mention two popular approaches. Our purpose is not to evaluate or advocate any of these approaches, since they are not the object of this study. Our purpose here is to open up space for migration studies scholars and policymakers to engage in the discussion.

One popular approach that could enhance human rights practices is “naming and shaming”. In short, members of the international society, including non-state organizations such as Human Rights Council and Amnesty International, can publicly condemn human rights abusers and pressurize them to change their abusive practices. There is supportive evidence for the effectiveness of this approach (e.g. Franklin, 2008). However, naming and shaming assumes that violators would observe the norms established in the international society to various extents and change their behaviours under the pressure of socialization (Finnemore and Sikkink, 1998). It also requires different actors to respond to the violations and condemn abuses co-ordinately. Consequently, this strategy works well when there is a strong civil society, high media attention and a large number of international non-governmental human rights organizations (INGOs) in a country (Neumayer, 2005; Meernik et al., 2012; Murdie and Davis, 2012). The naming and shaming strategy is more likely to be effective when complemented with other measures such as sanctions to increase the costs of human rights abuses (Lebovic and Voeten, 2009).

Another approach is to use economic incentives as leverage to promote human rights (Hafner-Burton, 2005; Marx et al., 2016, 2017). The idea behind this is that countries and international organisations could tie material benefits to compliance by integrating human rights standards into an international agreement. This will give potential human rights abusers greater incentives for compliance. The European Union (EU), for example, has put this into practice. The Lomé Convention, first signed in 1975 and expired in 2000, was one of the trade and aid cooperation between the European Economic Community (EEC) and African, Caribbean, and Pacific (ACP) states. It tried to condition development assistance on the human rights condition of the ACP states (Moravcsik, 1995; Hurt, 2003). In this way, ACP states were incentivized to improve their human rights records. Another example is the EU's Everything But Arms (EBA) arrangement, which grants an external economy access to the European single market for all products except arms and ammunition, free of quotas and tariffs. In October 2018, the EU stated that they were reviewing Cambodia's duty-free access to the single market due to the deteriorating human rights condition in the country (European Commission, 2018).

Cross-country analysis by Hafner-Burton (2005) confirms that states which enter PTAs with high standards performed better in their human rights protection practices. Recently, the EU has established a new architecture of international labour standards within the Trade and Sustainable Development (TSD) chapters of its trade agreements with the Caribbean countries, South Korea and Moldova, although Harrison et al. (2018) found no evidence that TSD chapters led to improvements in labour standards. Other similar examples exist. For instance, Vogt (2015) found that US managed to use the leverage created by the negotiation of trade agreements to pressure some of her trading partners to introduce reforms to laws related to labour's rights. Research by Bush (2011) provided yet another policy option. She found that the provision of foreign aid can affect the likelihood of a country to adopt a gender quota system that improves women's rights in a country.

Conclusion

This study showed that low respect for human rights, which includes physical integrity as well as political liberty, is a powerful driving force that motivates an individual to adopt an “exit” strategy. Although previous analyses showed that factors such as income, historical legacy and violent conflict are important determinants of migration, we find that the human rights situation of a country, a usually overlooked factor, is another key driver of the migration decision. This study goes beyond the existing physical integrity approach to defining human rights. Migrants are influenced by the level of freedom and liberty that they can enjoy in their countries of origin. They seek destinations that can maximize their welfare with minimum constraints. The aspirations to improve their living and to thrive motivate people to migrate. In some cases, migration can also be regarded as an exit strategy as well as an insurance that guards against harm due to lack of protection. Using bilateral data and controlling for multilateral resistances and sample selection bias, our analysis found supportive evidence for the hypothesis. Our finding implies that policy efforts to promote human rights could also be assessed based on their contribution to migration flows. As long as global migration produces negative spill-over effects to the receiving countries, migration is not only a domestic problem but a foreign policy issue.

Acknowledgements

We would like to thank the anonymous referees for their valuable comments and suggestions, especially the Everything But Arms example, from one of the referees.

References

Abusharaf, R.M. 1997 “ Sudanese migration to the new world: Socio-economic characteristics”, International Migration, 35(4): 513– 536.
Wiley Online LibraryCASPubMedWeb of Science®Google Scholar
Anderson, J.E., and E. van Wincoop 2003 “ Gravity with gravitas: A solution to the border puzzle”, American Economic Review, 93(1): 170– 192.
CrossrefWeb of Science®Google Scholar
Ariu, A., F. Docquier, and M.P. Squicciarini 2016 “ Governance quality and net migration flows”, Regional Science and Urban Economics, 60: 238– 248.
CrossrefWeb of Science®Google Scholar
Baier, S.L., and J.H. Bergstrand 2009 “ Bonus vetus OLS: A simple method for approximating international trade-cost effects using the gravity equation”, Journal of International Economics, 77(1): 77– 85.
CrossrefWeb of Science®Google Scholar
Baudassé, T., R. Bazillier, and I. Issifou 2018 “ Migration and institutions: Exit and voice (from abroad)?”, Journal of Economic Surveys, 32(3): 727– 766.
Wiley Online LibraryWeb of Science®Google Scholar
Beckerman, W. 1956 “ Distance and the pattern of intra-European trade”, Review of Economics and Statistics, 38(1): 31– 40.
CrossrefWeb of Science®Google Scholar
Beine, M., S. Bertoli, and J. Fernandez-Huertas Moraga 2016 “ A practitioner's guide to gravity models of international migration”, World Economy, 39(4): 496– 512.
Wiley Online LibraryWeb of Science®Google Scholar
Bertoli, S., and J.F.-H. Moraga 2013 “ Multilateral resistance to migration”, Journal of Development Economics, 102: 79– 100.
CrossrefWeb of Science®Google Scholar
Bilodeau, A. 2008 “ Immigrants’ voice through protest politics in Canada and Australia: Assessing the impact of pre-migration political repression”, Journal of Ethnic and Migration Studies, 34(6): 975– 1002.
CrossrefWeb of Science®Google Scholar
Bohra-Mishra, P., and D.S. Massey 2011 “ Individual decisions to migrate during civil conflict”, Demography, 48(2): 401– 424.
CrossrefPubMedWeb of Science®Google Scholar
Boyd, M. 1989 “ Family and personal networks in international migration: Recent developments and new agendas”, International Migration Review, 23(3): 638– 670.
CrossrefCASPubMedWeb of Science®Google Scholar
Bush, S.S. 2011 “ International politics and the spread of quotas for women in legislatures”, International Organization, 65(1): 103– 137.
CrossrefWeb of Science®Google Scholar
Carling, J. 2002 “ Migration in the age of involuntary immobility: Theoretical reflections and Cape Verdean experiences”, Journal of Ethnic and Migration Studies, 28(1): 5– 42.
CrossrefGoogle Scholar
Carroll, C.D., K.E. Dynan, and S.D. Krane 2003 “ Unemployment risk and precautionary wealth: Evidence from households’ balance sheets”, Review of Economics and Statistics, 85(3): 586– 604.
CrossrefWeb of Science®Google Scholar
Charvet, J., and E. Kaczynska-Nay 2008 The Liberal Project and Human Rights: The Theory and Practice of a New World Order. Cambridge University Press, Cambridge.
CrossrefGoogle Scholar
Cholewinski, R.I. 1997 Migrant Workers in International Human Rights Law: Their Protection in Countries of Employment. Clarendon Press, Oxford.
Web of Science®Google Scholar
Cingranelli, D.L., and D.L. Richards 2010 “ The Cingranelli and Richards (CIRI) human rights data project”, Human Rights Quarterly, 32(2): 401– 424.
CrossrefWeb of Science®Google Scholar
Cole, W.M. 2015 “ Mind the gap: State capacity and the implementation of human rights treaties”, International Organization, 69(2): 405– 441.
CrossrefWeb of Science®Google Scholar
Crush, J. 2001 “ The dark side of democracy: Migration, xenophobia and human rights in South Africa”, International Migration, 38(6): 103– 133.
Wiley Online LibraryWeb of Science®Google Scholar
De Haas, H. 2010 “ Migration and development: A theoretical perspective”, International Migration Review, 44(1): 227– 264.
Wiley Online LibraryPubMedWeb of Science®Google Scholar
Dustmann, C., and F. Fabbri 2003 “ Language proficiency and labour market performance of immigrants in the UK”, Economic Journal, 113(489): 695– 717.
Wiley Online LibraryWeb of Science®Google Scholar
European Commission 2018 “ Cambodia: EU mission assesses human rights and labour situation” http://trade.ec.europa.eu/doclib/press/index.cfm?id=1889 (accessed 10 December 2018).
Google Scholar
Fariss, C.J. 2014 “ Respect for human rights has improved over time: Modeling the changing standard of accountability”, American Political Science Review, 108(2): 297– 318.
CrossrefWeb of Science®Google Scholar
Finnemore, M., and K. Sikkink 1998 “ International norm dynamics and political change”, International Organization, 52(4): 887– 917.
CrossrefWeb of Science®Google Scholar
Franklin, J.C. 2008 “ Shame on you: The Impact of human rights criticism on political repression in Latin America”, International Studies Quarterly, 52(1): 187– 211.
CrossrefWeb of Science®Google Scholar
Gallagher, A. 2002 “ Tracking, smuggling and human rights: Tricks and treaties”, Forced Migration Review, 12(25): 8– 36.
Google Scholar
Gibney, M., and M. Dalton 1996 “ The political terror scale”, in D.L. Cingranelli (Ed.), Human Rights and Developing Countries. JAI Press, Greenwich: 73– 84.
Google Scholar
Greenwood, M.J., G.L. Hunt, and J.M. McDowell 1986 “ Migration and employment change: Empirical evidence on the spatial and temporal dimensions of the linkage”, Journal of Regional Science, 26(2): 223– 234.
Wiley Online LibraryCASPubMedWeb of Science®Google Scholar
Hafner-Burton, E.M. 2005 “ Trading human rights: How preferential trade agreements influence government repression”, International Organization, 59(3): 593– 629.
CrossrefWeb of Science®Google Scholar
Harris, J.R., and M.P. Todaro 1970 “ Migration, unemployment and development: A two-sector analysis”, American Economic Review, 60(1): 126– 142.
Web of Science®Google Scholar
Harrison, J., M. Barbu, and L. Campling, et al. 2018 “ Governing about standards through free trade agreements: Limits of the European Union's Trade and Sustainable Development Chapters”, Journal of Common Market Studies. https://doi.org/10.1111/jcms.12715
Wiley Online LibraryWeb of Science®Google Scholar
Hathaway, J.C. 1991 “ Reconceiving refugee law as human rights protection”, Journal of Refugee Studies, 4(2): 113– 131.
CrossrefGoogle Scholar
Hatton, T.J. 2009 “ The rise and fall of asylum: What happened and why?”, Economic Journal, 229(535): F183– F213.
Wiley Online LibraryWeb of Science®Google Scholar
Head, K., T. Mayer, and J. Ries 2010 “ The erosion of colonial trade linkages after independence”, Journal of International Economics, 81(1): 1– 14.
CrossrefWeb of Science®Google Scholar
Heckman, J.J. 1979 “ Sample selection bias as a specification error”, Econometrica, 47(1): 153– 161.
CrossrefWeb of Science®Google Scholar
Hugo, G. 1996 “ Environmental concerns and international migration”, International Migration Review, 30(1): 105– 131.
CrossrefCASPubMedWeb of Science®Google Scholar
Hurt, S.R. 2003 “ Co-operation and coercion? The Cotonou Agreement between the European Union and ACP states and the end of the Lomé Convention”, Third World Quarterly, 24(1): 161– 176.
CrossrefWeb of Science®Google Scholar
Ibáñez, A.M., and C.E. Vélez 2008 “ Civil conflict and forced migration: The micro determinants and welfare losses of displacement in Colombia”, World Development, 36(4): 659– 676.
CrossrefWeb of Science®Google Scholar
Jones, R.C. 1989 “ Causes of Salvadoran migration to the United States”, Geographical Review, 79(2): 183– 194.
CrossrefWeb of Science®Google Scholar
Karemera, D., V.I. Oguledo, and B. Davis 2000 “ A gravity model analysis of international migration to North America”, Applied Economics, 32(13): 1745– 1755.
CrossrefWeb of Science®Google Scholar
Keith, L.C. 1999 “ The United Nations International Covenant on Civil and Political Rights: Does it make a difference in human rights behavior?”, Journal of Peace Research, 36(1): 95– 118.
CrossrefWeb of Science®Google Scholar
Kim, H., and K. Sikkink 2010 “ Explaining the deterrence effect of human rights prosecutions for transitional countries”, International Studies Quarterly, 54(4): 939– 963.
CrossrefWeb of Science®Google Scholar
Lebovic, J.H., and E. Voeten 2009 “ The cost of shame: International organizations and foreign aid in the punishing of human rights violators”, Journal of Peace Research, 46(1): 79– 97.
CrossrefWeb of Science®Google Scholar
Marx, A., B. Lein, and N. Brando 2016 “ The protection of labour rights in trade agreements: The case of the EU-Colombia agreement”, Journal of World Trade, 50(4): 587– 610.
Web of Science®Google Scholar
Marx, A., F. Ebert, and N. Hachez 2017 “ Dispute settlement for labour provisions in EU Free Trade Agreements: Rethinking current approaches”, Politics and Governance, 5(4): 49– 59.
CrossrefWeb of Science®Google Scholar
Massey, D.S. 1990 “ The social and economic origins of immigration”, Annals of the American Academy of Political and Social Science, 510: 60– 72.
CrossrefWeb of Science®Google Scholar
Mayda, A.M. 2010 “ International migration: A panel data analysis of the determinants of bilateral flows”, Journal of Population Economics, 23(4): 1249– 1274.
CrossrefWeb of Science®Google Scholar
McKenzie, D., C. Theoharides, and D. Yang 2014 “ Distortions in the international migrant labor market: Evidence from Filipino migration and wage responses to destination country economic shocks”, American Economic Journal: Applied Economics, 6(2): 49– 75.
CrossrefWeb of Science®Google Scholar
McLeman, R., and B. Smit 2006 “ Migration as an adaptation to climate change”, Climate Change, 76(1–2): 31– 53.
CrossrefWeb of Science®Google Scholar
Meernik, J., R. Aloisi, M. Sowell, et al. 2012 “ The impact of human rights organizations on naming and shaming campaigns”, Journal of Conflict Resolution, 56(2): 233– 256.
CrossrefWeb of Science®Google Scholar
Moore, W.H., and S.M. Shellman 2004 “ Fear of persecution: Forced migration, 1952-1995”, Journal of Conflict Resolution, 48(5): 723– 745.
CrossrefWeb of Science®Google Scholar
Moravcsik, A. 1995 “ Explaining international human rights regimes: Liberal theory and Western Europe”, European Journal of International Relations, 1(2): 157– 189.
CrossrefGoogle Scholar
Morrison, A.R., and R.A. May 1994 “ Escape from terror: Violence and migration in post-revolutionary Guatemala”, Latin American Research Review, 29(2): 111– 132.
Web of Science®Google Scholar
Murdie, A.M., and D.R. Davis 2012 “ Shaming and blaming: Using events data to assess the impact of human rights INGOs”, International Studies Quarterly, 56(1): 1– 16.
CrossrefWeb of Science®Google Scholar
Nankani, G., J. Page, and L. Judge 2005 “ Human rights and poverty reduction strategies: Moving towards convergence?”, in P. Alston and M. Robinson (Eds.), Human Rights and Development: Towards Mutual Reinforcement. Oxford University Press, Oxford: 475– 497.
CrossrefGoogle Scholar
Narayan, P.K., and R. Smyth 2005 “ Temporal causality and the dynamics of democracy, emigration and real income in Fiji”, International Review of Applied Economics, 19(2): 245– 261.
CrossrefGoogle Scholar
National Bureau of Statistics of China (NBSC) 2015 China Statistical Yearbook 2014. China Statistics Press, Beijing.
Google Scholar
Neumayer, E. 2005 “ Do international human rights treaties improve respect for human rights?”, Journal of Conflict Resolution, 49(6): 925– 953.
CrossrefWeb of Science®Google Scholar
Ortega, F., and G. Peri 2013 “ The effect of income and immigration policies on international migration”, Migration Studies, 1(1): 47– 74.
CrossrefWeb of Science®Google Scholar
Özden, C., C.R. Parsons, M. Schiff, et al. 2011 “ Where on earth is everybody? The evolution of global bilateral migration 1960–2000”, World Bank Economic Review, 25(1): 12– 56.
CrossrefWeb of Science®Google Scholar
Palloni, A., D.S. Massey, M. Ceballos, et al. 2001 “ Social capital and international migration: A test using information on family networks”, American Journal of Sociology, 106(5): 1262– 1298.
CrossrefWeb of Science®Google Scholar
Pence, K.M. 2006 “ The role of wealth transformation: An application to estimating the effect of tax incentives on saving”, Contributions to Economic Analysis and Policy, 5(1): 1– 24.
PubMedGoogle Scholar
Pesaran, M.H. 2006 “ Estimation and inference in large heterogeneous panels with a multifactor error structure”, Econometrica, 74(4): 967– 1012.
Wiley Online LibraryWeb of Science®Google Scholar
Plant, V. 2005 “ Honor killings and the asylum gender gap”, Journal of Transnational Law and Policy, 15: 109– 129.
Google Scholar
Portela, C. 2010 European Union Sanctions and Foreign Policy: When and Why do They Work?. Routledge, New York.
Google Scholar
Radnitz, S. 2006 “ Weighing the political and economic motivations for migration in post-soviet space: The case of Uzbekistan”, Europe-Asia Studies, 58(5): 653– 677.
CrossrefWeb of Science®Google Scholar
Santos Silva, J.M.C.S., and S. Tenreyro 2006 “ The log of gravity”, Review of Economics and Statistics, 88(4): 641– 658.
CrossrefWeb of Science®Google Scholar
Schmeidl, S. 1997 “ Exploring the causes of forced migration: A pooled time-series analysis, 1971-1990”, Social Science Quarterly, 78(2): 284– 308.
Web of Science®Google Scholar
Semykina, A., and J.M. Wooldridge 2010 “ Estimating panel data models in the presence of endogeneity and selection”, Journal of Econometrics, 157(2): 375– 380.
CrossrefWeb of Science®Google Scholar
United Nations General Assembly 2013 Report of the special rapporteur on extrajudicial, summary or arbitrary executions. A/HRC/23/47/Add.4.
Google Scholar
United Nations Population Division 2015 Trends in International Migrant Stock: Migrants by Destination and Origin: The 2015 Revision. United Nations Database, POP/DB/MIG/Stock/Rev.2015.
Google Scholar
Vogt, J.S. 2015 “ The evolution of labor rights and trade—A transatlantic comparison and lessons for the transatlantic trade and investment partnership”, Journal of International Economic Law, 18(4): 827– 860.
CrossrefWeb of Science®Google Scholar
Wood, R.M., and M. Gibney 2010 “ The political terror scale (PTS): A re-introduction and a comparison to CIRI”, Human Rights Quarterly, 32(2): 367– 400.
CrossrefWeb of Science®Google Scholar
Wooldridge, J.M. 1995 “ Selection corrections for panel data models under conditional mean independence assumptions”, Journal of Econometrics, 68(1): 115– 132.
CrossrefWeb of Science®Google Scholar
World Bank 2017 “ World development indicators”, https://data.worldbank.org/data-catalog/world-development-indicators.
Google Scholar
World Bank 2018 “ Net migration”, https://data.worldbank.org/indicator/SM.POP.NETM?locations=CN.
Google Scholar
Zweig, D. 2006 “ Learning to compete: China's efforts to encourage a reverse brain drain”, in C. Kuptsch and E.F. Pang (Eds.), Competing for Global Talent. International Labour Organization, Geneva, Switzerland.
Google Scholar

Citing Literature

Volume57, Issue3

June 2019

Pages 98-114

Human Rights, Income and International Migration

Abstract

Introduction

Migration and Human Rights