Abstract
Morocco, once primarily known as a country of emigration and transit to Europe, has become a destination country for migrants, the majority of whom are from sub-Saharan Africa. Using an original nationally representative survey of 2,700 respondents, together with data from two census waves and one migrant regularization wave, the authors examine Moroccan citizens’ labor market threat perceptions from this new migrant group. Consistent with findings from studies conducted in developed countries, less educated, poorer respondents express higher labor market threat perceptions. The authors also find evidence, however, that city dwellers and employed female Moroccans are the most likely to report threat perceptions, even after controlling for greater “exposure” to migrants. The article contributes to literatures on migration exposure and how rural–urban dynamics shape labor market threat perceptions in a developing country.
North Africa, a subregion of the broader Arab world, has traditionally been a source of emigration. Indeed, between 5% and 10% of Moroccans, Tunisians, and Algerians reside abroad, mostly in Western Europe. North Africans abroad send remittances back home, providing major revenue for their families and also state tax coffers. Recently, however, North Africa has evolved into a destination for migrants (Norman 2016). Reports estimate that 2.5 million migrants and refugees actively live in North Africa, and most of them are sub-Saharan Africans (Huggler 2017: 4). The flow of sub-Saharan migrants into North Africa has the potential to dramatically alter the dynamics of local labor markets that are already burdened by high unemployment, informality, and rural–urban migration. North African citizens—similar to citizens of Europe and North America now hosting foreign migrants and refugees—may have anxieties about how the arrival of large numbers of sub-Saharan migrants could negatively affect their economic livelihoods in the labor market. Yet, little is known about the nature of labor market threat perceptions that North Africa’s host populations hold toward these migrants.
This study examines this topic in the case of Morocco, a North African society that currently hosts hundreds of thousands of migrants or refugees, with more than 80% originating from sub-Saharan African countries (Puschmann 2011: 22–23; Natter 2018: 12). Leveraging an original, nationally representative public opinion poll of 2,700 respondents, we document Moroccans’ labor market threat perceptions with respect to sub-Saharan migrants. Combining our survey data with data from two census waves and one migration regularization wave, we ask: What drives Moroccan citizens’ labor market threat perceptions from sub-Saharan migrants? Although some results from our survey confirm findings from studies pertaining to developed countries, we also find unique differences that we posit likely derive from the structure of Morocco’s labor market and the country’s dynamics of internal rural-to-urban migration. We make a theoretical contribution to the literature on migration “exposure” and the determinants of labor market threat perceptions in a developing country context.
We also make an empirical contribution, insofar as previous studies on migration and labor markets in the Middle East and North Africa (MENA) often take migrants themselves as the object of study. They use surveys, qualitative interviews, or ethnography to record migrants’ motivations, their intended destinations, and their hardships. Pioneering studies by Bachelet (2018, 2019a: 40, 2019b), Berriane (2015), and others have done this for sub-Saharan migrants in Morocco. A less developed strand in the literature focuses on host populations. Unlike studies from Western Europe and North America, which have extensively examined the views of polities receiving migrants, studies from the MENA that center on the attitudes of host populations are rare, and have focused primarily on polities receiving Syrian refugees (Getmansky, Sınmazdemir, and Zeitzoff 2018; Alrababa’h et al. 2021). Our novel contribution is that this study is one of the first of its kind from North Africa to systematically examine host populations’ perceptions about sub-Saharan migrants’ potential threats to the labor market. It also shows how these perceived labor market concerns manifest in a country characterized by high informality, low female labor market participation, and an ongoing process of urbanization.
Sub-Saharan African Migration to Morocco
Given instability and economic difficulties in sub-Saharan Africa in the 1990s, new waves of migrants and refugees came to Morocco, mainly seeing it, at least initially, as a transit country to Europe. González Enríquez (2007) noted that European elites during this period accused Morocco’s regime of “not being interested in preventing the departure of irregular sub-Saharan immigrants from its territory.” Indeed, permitting sub-Saharan migrants to depart for EU territory helped to not further exacerbate unemployment in Morocco’s labor market. Morocco, simply, “lacked incentives and financial and human resources to control this migration whose destination was Europe” (González Enríquez 2007: 228–29).
Under pressure from Spain, France, and the European Union, Morocco reversed its lax border enforcement in 1992. Morocco signed a bilateral treaty with Spain, wherein the latter agreed to pay for each migrant apprehended and repatriated to Moroccan territory (Alexander 2019a: 165). Complementary agreements with the European Union reinforced this policy and Morocco became one of North Africa’s first countries to act as a bulwark against sub-Saharan African migration to Europe (Schapendonk 2008: 133). This policy became known as border security externalization (Kent, Norman, and Tennis 2020; El Qadim et al. 2021; Buehler, Fabbe, and Kyrkopoulou 2022). Domestically, Morocco’s border externalization agreement manifested as Law 02-03, which granted the General Directorate for National Security sweeping powers to class migrants as “lawful or unlawful” (Filali Meknassi 2019). Morocco’s police became more violent, treating all sub-Saharan migrants “as if they were illegal” and ignoring differences between labor migrants and asylum seekers (Léonard 2010: 240–41; Natter 2018). Gazzotti (2021: 277) noted that police differentially imposed “pervasive containment procedures” on sub-Saharan migrants compared with non-Black migrants, even when their administrative statuses similarly contravened migration law. Sub-Saharan migrants often faced harsh workplace conditions, including as Bachelet (2014: 87) described, “precarious and poorly paid jobs” in the informal sector. Similarly, their living conditions were difficult, “ranging from overcrowded and hierarchical ‘ghettos’ to two people living in a room” (2014: 89).
These immigration policies tarnished Morocco’s image internationally and were met with various forms of collective resistance. Migrants, sometimes with support from local civil society organizations, created unions and associations to raise awareness about migrants’ demands for better working conditions, rights, and regularization (Üstübici 2016; Bachelet and Jeffery 2019; Norman 2021). One of the most notable associations, the Organisation Démocratique des Travailleurs Immigrés du Maroc, formed on May Day, 2012, and affiliates with one of Morocco’s most established, powerful labor unions, the Organisation Démocratique du Travail. Partially responding to such pressures, Morocco’s government softened its migration policies in 2013. Reforms included regularization campaigns in 2014 and 2017, which gave temporary legal status to 47,000 migrants (Alaoui 2018). Of regularized migrants, 90% were sub-Saharan Africans (Yabiladi 2018). This partial de-securitization of migration policies improved Morocco’s international reputation, helping its king promote himself as a “savior” of “vulnerable groups.” Similarly, the king signaled compassion by agreeing to host more Syrian refugees (Natter 2021: 861).
These changes in Morocco’s migration policies coincided with an improving domestic economy, both of which encouraged greater numbers of sub-Saharan migrants to remain in Morocco rather than seek passage to Europe. The result was the “de facto long-term settlement of sub-Saharan migrants in Morocco despite their labeling as transit migrants by governments” (Berriane, de Haas, and Natter 2015: 505). Morocco thus has evolved from a “transit” into a “destination” country (Bachelet 2019a; Norman 2020). In 2019, the Migration Policy Institute estimated 700,000 sub-Saharan African migrants resided in Morocco (Ghazouani 2019). Sub-Saharan migrants tend to live in larger “gateway” cities (e.g., Casablanca, Rabat) that have more work opportunities (Bachelet 2018). Despite regularization waves and the aforementioned moves toward softening migration policy, there have been many reports of continued violence against sub-Saharan migrants. A particularly tragic incident took place in June 2022 at the Melilla–Nador border, where clashes between migrants and border guards killed dozens of asylum seekers who were trying to gain entry into Spanish territory (AMDH 2022; GADEM 2022).
Although systematic data about the composition of sub-Saharan migrant groups in Morocco remains limited, its official statistics agency, the Haut-Commissariat au Plan (the HCP), conducted a 2021 survey of migrants residing in the country. This survey sampled 3,000 migrants, made up of 2,200 regularized or irregular migrants and 800 refugees or asylum seekers. Among all regularized or irregular migrants in the survey, an overwhelming majority (86.9%) were sub-Saharan Africans: 16.7% from Côte d’Ivoire, 15.9% from Senegal, 13.2% from Guinea, 10.1% from the Democratic Republic of Congo, 8.7% from Cameroon, 4.9% from Mali, 2.3% from the Central African Republic, and 15.1% from other sub-Saharan countries (HCP 2021: 5, 48). When asked about their primary reason for leaving their country, 39.8% of respondents said conflict and persecution in their country, 36.8% cited poor employment and living conditions, 14.2% identified few educational opportunities, and 4.7% reported family reunification. More than half (53.7%) of migrants surveyed plan to stay in Morocco, while only a quarter (25.8%) say they want to emigrate elsewhere (HCP 2021: 105).
Note that Moroccan citizens’ perceptions about sub-Saharan migrants’ potential labor market threats do not emerge in a vacuum. Rather, they are intertwined with the underlying legacy of Morocco’s racially stratified society (Menin 2020). In his pioneering book, Black Morocco: A History of Slavery, Race, and Islam, Chouki El Hamel describes the origins of this racial stratification. Black Moroccans—both Black Arabs and Black Amazigh—have existed in Morocco for centuries. Some arrived with Arab armies during the Islamic conquest of the early 8th century, while others settled in Morocco’s southern regions—primarily Drâa-Tafilalet and Souss-Massa—searching for better farmland (El Hamel 2013: 39). Many of these Black Moroccans had lived in Morocco “since time immemorial” and had “always been free” (El Hamel 2002: 39). Numerous Black Moroccans also arrived as enslaved peoples from West and Central Africa, as part of what Lydon (2005) described as the trans-Saharan trade route. These individuals are known as Ḥarāṭīn and are considered descendants of freed slaves who had once worked as soldiers, domestic workers, and in other roles (El Hamel 2013: 90–91). The term Ḥarāṭīn, roughly translated as “free men,” is often used to refer to Black Moroccans. Yet El Hamel (2013: 4–5) and Menin (2020: 166) noted its inaccuracy, as some Black Moroccans are not descendants of formerly enslaved peoples. Slavery, as El Hamel explains, was a “central component of economic systems” in pre-colonial Morocco and constructed a “foundation for a society divided first by skin color, and then by race” (2002: 30). Drawing on interviews with Senegalese migrants in Morocco, Menin (2016, 2020) showed how, until today, “black people (often independent of their ancestry) inherit the status of social and moral inferiority accorded to black slaves” (2020: 182).
Morocco’s Economy and Labor Market
Morocco has experienced economic progress over the past 20 years. Income per capita doubled between 2000 and 2019, health and literacy outcomes improved, access to water and electricity increased, and poverty fell to one-third of its 2000 level (Lopez-Acevedo, Betcherman, Khellaf, and Molini 2021). Phosphate and agricultural exports, the boom in competitive services, and the rapidly modernizing industrial sector have kept Morocco’s growth stable and mostly above average for North African countries (African Union Commission and OECD 2018). Morocco’s 2017 re-entrance into the African Union and its application to the Economic Community of West African States evidence efforts to enhance economic exchange with sub-Saharan Africa.
Between 2000 and 2020, a shift from the agricultural sector to services caused a notable change in the structure of Morocco’s labor market. In 2000, agriculture, industry, and services accounted for 45%, 20%, and 35% of employment, respectively. By 2020, the shares of employment had shifted to 34% for agriculture, 22% for industry, and 44% for services (Danish Trade Union Development Agency 2022). Despite these shifts, Morocco lags behind other middle-income countries in structural change in employment, given its remaining and relatively large share of agricultural employment (Lopez-Acevedo, Betcherman, et al. 2021: 17).
Regarding employment trends, per HCP 2021 data, 39.7% of Moroccans are employed and the unemployment rate is 12.3% (HCP 2022b). Morocco’s share of economically active people is 45.3%, with 20.9% of women and 70.4% of men (HCP 2022c). The employment rate is much lower for women, standing at only 22.3% and indicating a long-standing lack of female labor force participation (HCP 2022f). Also, unemployment is much higher for younger (31.8% for ages 15–24; 19.6% for ages 25–34) and urban-residing Moroccans (16.9% for urban, 5.0% for rural) (HCP 2022e). Urban unemployment for those aged 15 to 24 is especially high, at 45.3%, signaling a particularly challenging issue of youth unemployment in cities (HCP 2022d). Morocco has a low share of public-sector employment (8.7%), remaining lower than all MENA countries as well as the global trend line (HCP 2022a: Hertog 2023: 29).
Further, Morocco has high levels of labor informality. Islam, Moosa, and Saliola (2022) reported that, in 2019, 77% of total employment in Morocco was informal, the highest in the MENA. The share of informal employment is even higher in the private sector (84%), since most private jobs lack pensions and other social insurance. Of note, Moroccans do not express high expectations for governmental social insurance and welfare programs, making them a regional exception (Hertog 2023: 47). Holland (2017) found that sometimes governments may allow for the violation of their own labor laws—tolerating, for example, street vending and various kinds of informal work—in order to win the poor’s support. In this sense, “forbearance” of labor law becomes “an informal welfare policy” (Holland 2017). This may also be true for Morocco, given its large informal sector.
Finally, Morocco has experienced considerable internal migration of citizens from rural to urban areas, shaping the structure of its labor market. Casablanca has historically drawn large numbers of migrants from Morocco’s less developed rural interior for industrial jobs (Puschmann 2011: 151–52). These rural–urban migration trends have intensified over time, with approximately 29.3% of Moroccans living in cities in the 1960s, to approximately 48.4% in the 1990s, to more than 64% in 2021 (African Population Newsletter 1995: 5). Karam and Decaluwé explained that this rural–urban migration stems from “climatic risks associated with agricultural production that encourage farmers to look for stable employment in urban areas” (2010: 499). Their rough estimates indicated that approximately 200,000 Moroccans were moving into urban areas annually in the early 2000s (2010: 499). Between 2000 and 2019, urban population gains represented 97% of the total population increase in Morocco (Lopez-Acevedo, Betcherman, et al. 2021: 18). Ultimately, although rural migrants initially “came to find jobs in the industrial sector,” as Puschmann concluded, their swelling numbers have generated “ominous signs of over-urbanization” including high urban unemployment and shantytowns (Puschmann 2011: 152).
Skills and Occupations of Sub-Saharan Migrants in Morocco
Sub-Saharan migrants arrive in Morocco with a variety of professional skill sets and education levels. According to secondary literature, common professions for migrants with lower skills include shoemakers, security guards, and laborers in masonry or construction work (Karam and Decaluwé 2010: 503). Others often work in the informal sector as unlicensed street vendors at marketplaces, selling household items, fruits, used clothing, and other goods (Alexander 2019b: 794). Sub-Saharan migrants with higher skills seek jobs in Morocco’s French-language press, telecommunication companies and call centers, and tile and marble factories (Berriane 2015: 586; Cherti and Collyer 2015: 595). Some high-skilled migrants arrive legally as public university students and then stay to work after their period of legal residency for schooling lapses (Berriane 2015: 586). Several studies noted the tendency of sub-Saharan migrants to cluster in the informal labor market, though this is difficult to measure (Karam and Decaluwé 2010: 504; Cherti and Collyer 2015: 598).
The aforementioned HCP report offers more information on the educational attainment, employment status, and income sources of migrants residing in Morocco. With respect to educational background, the survey found that nearly a quarter of migrants finished high school (23.5%) and more than another quarter acquired tertiary education (27.3%). Almost a third of migrants received additional vocational training, 17.3% in their country of origin and 13.9% in Morocco. The HCP survey does not provide data on the degree to which these skills and qualifications are used in the Moroccan labor market, but evidence suggests that they are underused, similar to most labor markets of southern Mediterranean countries (European Training Foundation 2021). More than half of migrants were employed in their home country in the three months preceding their migration to Morocco, while 11.5% were unemployed, and more than a third were outside the labor market.
Concerning migrants’ employment in Morocco, the HCP’s survey documents that close to half of migrants are employed (48%), whereas just over a quarter (27.4%) are unemployed and another quarter (24.5%) remain outside the labor market. Migrant employment is concentrated in commerce and services: 53% of employed migrants report working in services, 22% in trade, 12.8% in the construction sector, 7.9% in agriculture, and 4.3% in industry. The percentage of employed migrants increases relative to their education level. While 32% of migrants without formal education report having a job, this figure raises to 47.7% for those with a high school degree and to 58.1% for those with tertiary education. Per the HCP survey, the employment sectors in which sub-Saharan migrants work vary by their home country. More Central Africans (81.2%), DRC nationals (71.7%), and Cameroonians (66.7%) work in the service sector, compared with Senegalese (34.9%) and Guineans (35.4%). Trade occupies relatively more Senegalese (38.8%), Malians (27.8%), and Ivorians (25.8%), while more Guineans (30.1%) work in construction. More Malians (24.3%), and to a lesser extent Senegalese (15%), work in agriculture (HCP 2021: 90).
Theory: Moroccans’ Labor Market Threat Perceptions of Sub-Saharan Migrants
Against this backdrop, literature from studies of Western Europe and North America (Alesina and Tabellini 2022; Fabbe, Kyrkopoulou, and Vidali 2022) and also from Morocco (cited below) help us theorize how citizens perceive three potential labor market threats from migrants—job competition, aggregate wage reductions, and concerns about migrants’ skills.
Citizens’ Perceptions That Migrants “Steal” Jobs
First, Morocco’s citizens may perceive that sub-Saharan migrants take their jobs. Classic surveys of Western European publics found that citizens fear that new migrants undermine their economic security through competition over employment (Scheve and Slaughter 2001; Mayda 2006; O’Rourke and Sinnott 2006). The basic logic is that because more workers enter the labor market, yet the number of jobs does not expand, competition over employment intensifies. Empirically speaking, researchers have found mixed results as to whether migrants actually hurt citizens’ employment prospects. Whereas some studies found that migrants do negatively affect citizens’ employment prospects (Dustmann, Schönberg, and Stuhler 2017), others found that they do not have much of a negative effect (Cengiz and Tekgüç 2022), or that they negatively affected only those who enter the low-skill labor market during a large immigration influx (Monras 2020). Still, others found mixed effects based on profession and geographic location. For example, Card (2001) found no negative effect on employment, except for in “gateway cities” where migrants arrive and enter low-skilled professions. Lastly, Tabellini (2020) explored European immigration to US cities between 1910 and 1930, finding that immigrants actually increased native workers’ employment prospects by fostering greater industrial production.
Regardless of whether migrants actually outcompete native workers for jobs, perceptions exist among some citizens that this dynamic is real. These perceptions of job competition especially seem to cluster among poorer, low-skilled citizens, who tend to work the same jobs as migrants (Mayda 2006). Such perceptions of job competition from migrants also intensify among citizens who are unemployed or who see the national economy as unhealthy (Semyonov, Raijman, and Gorodzeisky 2006). Furthermore, O’Rourke and Sinnott (2006) found that high-skilled workers are less opposed to immigrants, and this effect is stronger in richer and more equal countries.
Morocco-based ethnographic studies found that some citizens perceive sub-Saharan migrants as taking jobs. Some Moroccans, as Alexander (2019b) noted, simply see migrants as harder workers. One of her interviewees remarked: I have a friend who runs a construction company, and he pays Africans 20 dirhams an hour [about $2.25] and Moroccans 10 dirhams an hour [about $1.15], because he knows that Africans work twice as hard. Africans are hard working. If Moroccans worked harder, he would pay them more. They’re going to take all of our jobs from us soon.” (2019b: 790)
Another of Alexander’s interviewees selling goods in a marketplace similarly noted, “Every day, it seems like there’s one more migrant hawking here. There’s one less Moroccan” (Alexander 2019b: 794). Others suggest sub-Saharan migrants are perceived to be more willing to work difficult, informal, and low-paying jobs (Bachelet 2014: 87). Though it is difficult to know how representative these opinions are within Morocco’s general population, these interviews provide some evidence that some citizens do see sub-Saharan migrants as edging-out locals for jobs.
Citizens’ Perceptions That Migrants Reduce Wages
Second, citizens may perceive that sub-Saharan migrants reduce the aggregate wage level. The rationale is that because migrants often accept lower wages than citizens, their increasing prevalence in the labor market depresses wages for all workers (De New and Zimmermann 1994; Card 2001). Several reasons explain why migrants sometimes accept lower wages. One reason is that foreign migrants often hail from lower-income, developing countries or post-Communist societies, with lower living costs and smaller salaries (Scheve and Slaughter 2001; Mayda 2006; O’Rourke and Sinnott 2006). A second reason is that because migrants are new to the country or perhaps have uncertain legal statuses, they tend to assert themselves less in negotiations with employers, thus resulting in lower wages (Bonacich 1972).
Like research on competition with migrants for jobs, however, empirical studies found conflicting results as to whether migrants actually reduced citizens’ wages. Some studies emphasized that migrants do indeed create downward pressure on citizens’ wages (Aydemir and Borjas 2007). Others found that this downward pressure occurred mostly for the wages of poorer, low-skilled workers (Borjas 2003). Some studies found, conversely, that migrants did not depress citizens’ wages at all (Card 2005). Or, if they did, it was only a temporary phenomenon in the economic short-term (Monras 2020).
Relatedly, Ottaviano and Peri (2012) showed that in the case of the United States, immigrants and US-born workers within the same education-experience-gender group were not perfect substitutes because they chose different occupations and had different skills. Interestingly, they found a positive effect of immigration on the wages of US-born workers with at least a high school degree, but a small negative effect on wages of US-born workers without a high school degree. A study of Syrian migrants’ impact on the Turkish economy similarly found that native lower-skilled workers suffered small wage and employment losses after the Syrian migrant wave, but higher-skilled workers have seen gains (Cengiz and Tekgüç 2022). Reviewing 12 UK studies conducted between 2003 and 2018, the Migration Advisory Committee (2018) concluded that immigration had little effect on average wages. Some studies (Nickell and Saleheen 2015) found a small negative impact on average wages, whereas others (Dustmann, Frattini, and Preston 2013) found positive average effects.
Empirical facts aside, many citizens do believe that migrants reduce aggregate wage levels. The citizens most concerned about migrants depressing wages tend to be poorer (Scheve and Slaughter 2001). By contrast, citizens with higher education levels expressed less concern (Hainmueller and Hiscox 2010b).
Studies of sub-Saharan migrants in Morocco show that some Moroccans fear that migrants depress wages. One study found that “sub-Saharan immigration to cities raises pressure on the urban labor market,” which cultivates perceptions of salary depression among Moroccans (Karam and Decaluwé 2010: 498). Perceptions of wage depression come from the fact that, as de Haas (2008) noted, Moroccans understand sub-Saharan migrants to hail from “substantially poorer origin countries” and thus would be willing to accept lower wages. In reality, sub-Saharan migrants in Morocco exhibit diverse socioeconomic backgrounds, and often come from educated families of “moderate means” (Kirwin and Anderson 2018: 16). Nonetheless, Moroccan citizens often misperceive that they are poor, and thus would willingly accept lower pay.
Citizens’ Perceptions That Migrants Are Unskilled
Third, citizens may perceive sub-Saharan migrants as a labor market threat because they believe them to be unskilled, and thus unable to contribute usefully to the economy. Studies suggest that people may express concerns about the fiscal burden immigration may pose on public services (Hainmueller and Hiscox 2010a). These concerns include the use of public health and education institutions, all types of welfare assistance, and other government services such as fire protection, police, parks, and roads. In this sense, unskilled immigrants may be perceived as a substantial net burden on public finances, whereas high-skilled immigrants are viewed as net contributors. Moreover, some scholars find that influxes of migrants without useful skills can have negative effects on the overall economy, discouraging employers from investing in technology and thereby hindering innovation (Lewis 2011; Peri 2012; Bratti and Conti 2018; Bulat 2019).
Morocco’s citizens often perceive sub-Saharan migrants as unskilled, which is a stereotype, again, derived from the assumption that they originate from poorer, less developed societies (de Haas 2008). In truth, and as explained above, sub-Saharan migrants have a diverse skill profile, including both high- and low-skill workers (Pickerill 2011).
Based on how citizens are theorized to perceive these three possible labor market threats from sub-Saharan migrants, we articulate seven hypotheses:
Methods: An Original Survey from Morocco
In 2018, we completed a nationally representative survey of 2,700 adult Moroccan citizens. To the best of the research team’s knowledge, our survey was the largest ever conducted assessing the views of citizens in North Africa concerning migration. Over the course of 63 days, the research team collaborated with a Moroccan public opinion firm (Global for Survey and Consulting) that possesses extensive local experience conducting nationally representative public opinion polls. The firm first began administering surveys in 2005 and has implemented approximately 60 nationally representative polls. This public opinion organization possesses legal permits to exist as a research firm in Morocco and has served as the collaborating partner on many international survey projects completed in the country.
Our study’s sampling approach for the survey was similar to that of Benstead’s pathbreaking Morocco-based public opinion polls (Benstead 2014a, 2014b). The research team employed a multistage, stratified probability sampling design. Random selection methods were utilized at all sampling levels to obtain a randomized, nationally representative sample for all of Morocco (that featured a +/– 2.3 margin of error). The administrative level of “region” served as sampling strata. Developing a randomized sample for the survey means that any adult citizen in Morocco possessed an equal probability of selection for inclusion as a respondent in the survey. For each separate region, the lowest-level administrative districts—what are termed the commune (baladiyyah or muqāṭa`ah in Arabic)—were selected randomly. Thus, the sample incorporated a diverse variety of communes with varying levels of poverty and rural–urban status. Thereafter, housing blocks were selected randomly within each commune. The research team deployed survey enumerators (i.e., interviewers) who administered the survey questionnaire to respondents at their doorsteps with verbal informed consent (according to the institutional review board’s protocol). The enumerators employed the random walk method to select homes at which to administer the survey. The enumerators randomly selected interviewees among adults at each abode by using the Kish grid method. Only adults were surveyed, and with full anonymity. Interviewers, adopting a procedure developed by the Afrobarometer international polling initiative, alternated interviewing men and women to ensure the sample included a balance of respondents from both sexes. The sample featured a nearly equal division of 50% female and 50% male respondents. To ensure the survey was weighted for each community’s population size, the survey firm contracted a demographic statistician from Morocco’s Haut-Commissariat au Plan. Source data on the socioeconomic attributes and population sizes for each region and commune originated from the most recent national census.
The research team also calculated the survey’s response rate, adopting the formula utilized by the Arab Barometer IV in Algeria, Jordan, Lebanon, and Palestine (response/(response + refusals)). Our public opinion poll featured an 81.68% response rate employing this formula. In 2020, the survey firm we worked with conducted a similar survey of 1,688 respondents, financed by the UK Economic and Social Research Council. The firm achieved a response rate of 73.07%, which was similar to ours. Researchers who are working with data drawn from online surveys or polls conducted in the United States may view these response rates as high, but—as noted by Bush and Jamal (2015: 38)—they are normal for household surveys completed in Arab countries, which frequently feature response rates above 90%. Such high response rates for household surveys derive from a number of characteristics specific to Arab countries. Foremost, the region has been less subjected to systematic survey research, in comparison with other world regions, leading to less saturation of survey requests. Second, the region has high rates of male unemployment and low levels of female workforce participation. Households thus have fewer absences during working hours, facilitating higher survey response rates.
Before implementing the main survey, we tested the questionnaire with two separate pilot studies, administered in September 2017 and June 2018, respectively. The two pilot studies helped us polish the Arabic questions to improve their clarity and understandability for survey participants. The study’s Arabic questionnaire was translated by a professional Moroccan translator working in Morocco and familiar with the local dialect. One of this study’s authors, who is professionally fluent in Arabic, and also the survey organization’s Moroccan director (a native Arabic speaker) additionally reviewed the translation. Prior to deploying the survey questionnaire, the study’s authors and the director of the polling firm organized a training workshop for the survey teams. Over two days, the survey teams familiarized themselves with the questionnaire and practiced administering it to respondents. The survey interviewers were all native Arabic-speaking Moroccan citizens who possessed previous work experience administering surveys for the same polling company. In the field, all survey questionnaires were administered to respondents by tablet, and their answers were uploaded every day into the data set. The tablet recorded an automatic time and geospatial location stamp for each interview to track survey enumerators working in the field.
This survey was unlikely to have caused social desirability bias among respondents because the local media frequently covers issues of sub-Saharan migration, so this is a normal topic of discussion in daily life in Morocco. In addition, our survey had a low refusal rate (i.e., less than 1% refusal for the three main dependent variable questions). Had the respondents found the topic too sensitive, a larger refusal rate would likely have occurred. And finally, our universities’ institutional review committees approved the questionnaire, indicating that these questions were not too sensitive.
Dependent Variables for Empirical Test
This study centers on three dependent variables that track variation in Moroccan citizens’ perceptions of sub-Saharan migrants’ potential labor market threats. A codebook for all variable definitions is in the Online Appendix. At the doorstep of each respondent’s home, the survey enumerator began with a contextualizing descriptive statement: Many of the refugees and migrants who have come to Morocco are Black Africans from sub-Saharan African countries, like Nigeria, Congo, Senegal, Guinea-Conakry, Ivory Coast, Mali, Cameroon, and others. Some Moroccans think that Black African refugees and migrants are threats and cause problems in Morocco. Other Moroccans disagree and do not think they are threats. Next, I will ask you questions on this topic. These are only your personal opinions, and there are no correct or false answers.
Respondents were then asked whether they “Strongly Agree” (5), “Agree” (4), are “Neutral” (3), “Disagree” (2), or “Strongly Disagree” (1) with the following statements about sub-Saharan migrants. The first statement—dependent variable one—asked whether “Black African refugees and migrants threaten Morocco by taking jobs from Moroccans.” The second statement—dependent variable two—inquired whether “Black African refugees and migrants threaten Morocco by accepting lower wages than Moroccans, which reduces all wages in the labor market.” The third statement—dependent variable three—questioned whether “Black African refugees and migrants threaten Morocco because they are unskilled workers.”
We note that by 2020, two years after our survey’s implementation, a consensus emerged in English scholarship that “sub-Saharan migrants” is the best term for such individuals (Aïdi 2020). This study, in discussing relevant literature and hypotheses, thus utilizes “sub-Saharan migrants.” This term, however, is not widely known or used by Moroccans, who often describe this group as “Black African migrants.” Our pilot studies helped us determine that our study would benefit from using the term most understood by Moroccan respondents. We also used this term to ensure respondents did not misunderstand the survey to mean Arab migrants from sub-Saharan African countries, another point of possible confusion. Some Malians and Senegalese (about 10% and 0.5%, respectively) self-identify as ethnically Arab or mixed Arab/Amazigh. Mauritania, sometimes considered a sub-Saharan African country, has 30% mixed Arab/Amazigh citizens. The existence of Arab and Arab/Amazigh citizens of sub-Saharan countries, as well as indigenous Black Moroccans, provide evidence that boundaries between North and sub-Saharan Africa are somewhat arbitrary, and were reified during French colonialism (Hannoum 2021).
Independent Variables
The independent variables used in our study explore the attributes correlated with Moroccans theorized to be more (or less) likely to perceive potential labor market threats from migrants. First, we created independent variables accounting for the respondent’s level of wealth, education, skills, and employment status. Education shows the education level of the respondent, with higher values indicating a higher level of education. Later in the article we provide descriptive statistics of how the perceived labor market threats vary by the respondent’s level of education. Income level (low) is a binary variable that takes the value 1 if the respondent stated that they are poor, that is, their family income is not sufficient to cover their expenses and they face difficulties in covering their needs, and 0 otherwise. In our data, 73% of respondents who were poor reported that they are concerned or strongly concerned about sub-Saharan migrants threatening Morocco by taking jobs, compared to 60% concerned respondents in the overall sample. The respective percentage of poor respondents reporting wage concerns about migrants is even higher, reaching 76%, compared to 65% in the overall sample. And of these poor respondents, 67% are concerned about the economic impact of unskilled migrants, compared to 42% in the total survey sample. White collar is a binary variable if the respondent is in a white-collar profession (e.g., doctor, lawyer), and 0 otherwise. Employed is a binary variable that takes the value 1 if the respondent is currently employed, and 0 otherwise.
To explore the rural–urban dynamics, we included a variable recording urban residency and a variable capturing respondent’s internal migration status. Urban location is a binary variable that takes the value 1 if the respondent currently resides in an urban area, and 0 otherwise. Internal migration is a binary variable that takes the value 1 if the respondent grew up in a rural area but currently resides in an urban area, and 0 otherwise.
To account for immigration exposure, we incorporated three variables in the model. High sub-Saharan migrant presence is a binary variable, coded as 1 if the respondent resides in a province that has high sub-Saharan African immigrant presence according to the 2014 regularization wave, and 0 otherwise. Change in foreign population (%) captures the change in the foreign population percentage between the 2004 and the 2014 census in the respondent’s commune. Finally, Foreign population (%) is the overall foreign population percentage in the respondent’s commune, per the 2014 census. The Online Appendix further details the construction of the variables.
We also included control variables, such as a respondent’s age, gender, Black Moroccan background, Francophone status, and whether the respondent lives in an area with a historically high Black Moroccan presence. Female is a binary variable that takes the value 1 if the respondent is female, and 0 if the respondent is male. We also include an interaction term between the variables Employed and Female (Employed × Female) to capture any additional effects coming from employed female respondents. Age is the respondent’s reported age. French speaker is a binary variable that takes the value 1 if the respondent speaks mostly French with their family at home, at work, or in their studies, and 0 otherwise. Morocco is ethnically diverse, so we include a control variable recording the historically high presence of Black Moroccans in some regions. High Black Moroccan presence is a binary variable, which takes the value 1 if the region in which the respondent resides has historically had a large indigenous Black Moroccan community, and 0 otherwise. Black Moroccan respondent is a binary variable that takes the value 1 if the respondent is a Black Moroccan, and 0 otherwise.
To ensure our independent variables did not exhibit multicollinearity, we conducted a variable inflation factor (VIF) test and its maximum values fell below the traditional threshold of 5.0. Specifically, VIF tests were as follows: model 1 (1 to 2.40), model 2 (1 to 2.39), and model 3 (1 to 2.39). Table 1 shows summary statistics for the variables, Table 2 shows how perceived labor market threats vary by level of education, and Table 3 is the correlation matrix of independent variables.
Summary Statistics
Notes: Obs, observations; SD, standard deviation.
Descriptive Statistics
Notes: Table shows the percentage of those who agree or strongly agree that sub-Saharan migrants pose a labor market threat, by level of education of the respondent.
Correlation Matrix of Independent Variables
Results and Discussion
Our survey reveals that Moroccans are most concerned that sub-Saharan migrants would reduce aggregate wage levels by accepting lower pay (66% of respondents) or take their jobs (60%) but were less concerned that they were unskilled for the workforce (42%). We conduct z-tests on the equality of proportions between the samples and find that Moroccans’ concern about having their wages lowered due to sub-Saharan migrants accepting smaller salaries is statistically significantly higher than their fear of losing their jobs to them. Similarly, Moroccans’ anxiety about losing their job to a sub-Saharan migrant is significantly higher than their concern about sub-Saharan migrants being unskilled for the workforce. The tests are performed for no difference in population proportions estimated from the two samples. The p value of the z-test between the job threat and the wage threat is Pr(|Z| > |z|) = 0.000. The p value of the z-test between the wage threat and the unskilled threat is Pr(|Z| > |z|) = 0.000.
Results from the ordered logistic regressions bear out some of the seven hypotheses, whereas they do not support others. Table 4 presents these results, showing that certain types of Moroccans are statistically significantly more (or less) likely to perceive threats that sub-Saharan migrants might take their jobs, depress aggregate wages, and not contribute to the economy because of a lack of skills.
Labor Market Threat Perceptions
Notes: Standard errors in parentheses.
Statistically significant at the .10 level; ** at the .05 level; *** at the .01 level.
Our regression results support the first hypothesis (H1), as Moroccans with higher education levels are less likely to perceive possible negative labor market effects from sub-Saharan migrants. This result—consistent with existing studies—likely reflects that highly educated Moroccan workers are less likely to compete over the same jobs with such migrants. Put simply, educated respondents likely believe they have a low substitution rate with sub-Saharan migrants in the labor market. Results in Table 3 also confirm hypothesis two (H2) that the poorest citizens are more likely to perceive all three types of labor market threats from sub-Saharan migrants.
Next, we turn to hypothesis three (H3), which posited that unemployed respondents will be more likely to perceive labor market threats from migrants than those who are employed. Our results show that employment status is not correlated with higher reported labor market threats. But, given the gendered divisions in Morocco’s labor market, could employment’s effect on perceived threats differ by gender? To account for this, we add an interaction term between gender and employment status and find that employed female respondents are more likely to be concerned about losing their jobs to sub-Saharan migrants. Morocco exhibits a low female labor participation rate, and one that further declined from 2000 to 2018 (Lopez-Acevedo, Devoto, Morales, and Rodriguez 2021). In many places, women are stigmatized for holding manufacturing jobs (Caraway 2007). Simultaneously, female sectoral employment has shifted out of agriculture and into services and manufacturing (Lopez-Acevedo, Devoto, et al. 2021). These trends may indicate a self-selection of women who chose to enter the labor market, despite facing higher cultural and structural obstacles compared with men. Indeed, this finding provides an avenue for future research, which could explore the precise mechanisms behind why employed women feel more threatened by job competition with sub-Saharan migrants.
Further, our results strongly support hypothesis four (H4) that urban Moroccans are more concerned about sub-Saharan migrants’ perceived job competition, wage depression, and lower skills. This effect remains present while controlling for immigration exposure by region. The high flows of internal, rural–urban migration in Morocco raise the question of whether being an internal Moroccan migrant increases the perceived labor market threats from sub-Saharan migrants. We find this is not true. In fact, being a rural–urban internal migrant correlates with a lower probability of perceiving a threat from sub-Saharan migrants because they depress wages.
Moving on to immigration exposure, we find no support for H5 that respondents who reside in areas with high foreign immigrant presence are more likely to feel threatened. We only find partial support for H6, with respondents who reside in a region that experienced a large change in the foreign population being more likely to feel threatened regarding wage depression. We find no support for H7, indicating that a higher overall foreign population does not raise perceived labor market concerns.
We find no statistically significant effects for most demographic control variables—gender, age, or Black Moroccan background. Francophone respondents, however, are more likely to express higher labor market concerns from sub-Saharan migrants, which we discuss below. Additionally, respondents who live in areas with historically higher Black Moroccan presence are less likely to be concerned about sub-Saharan migrants taking their jobs, but more likely to be concerned that they are low-skilled.
Conclusion
Morocco has evolved from being a transit country to being a host country for migrants, the vast majority of whom come from sub-Saharan Africa. This shift in Morocco’s demographic structure has also changed conditions in the country’s labor market. Against the backdrop of these changes, we studied the determinants of Moroccans’ labor market threat perceptions from this migrant group. Consistent with research from other contexts, we find that educated Moroccans are less likely to perceive threats related to sub-Saharan migrants’ job competition, reduction of wages, or skill level. Poor Moroccans, by comparison, are more likely to perceive all three types of threats.
Our study begins to shed light on how citizens in the Global South view the increasingly large entry of migrants into their territories and also points to a number of potential avenues for future work. Our results appear to reflect existing rural–urban divisions in Morocco’s labor market, with urban respondents being more likely to perceive all three types of threats from migrants. We do not detect that these findings are driven by Morocco’s recent rural–urban internal migrants themselves, however, who, if anything, perceive less wage and job competition threats from migrants. Future studies could thus address whether incumbent, urban workers in Morocco see internal migrants and international migrants as substitutes, or whether other dynamics are at play. Additionally, we find that employed women are especially concerned about job competition from migrants. Further research could explore whether sectoral overlaps or some other feature of women’s still limited participation in Morocco’s labor force drive this finding. We find that French speakers, too, show heightened threat perceptions across all three categories. Given that the quantitative approach of our survey focused on “sub-Saharan” migrants as a generic category, it admittedly glosses over a wide range of diverse attributes that migrants might possess. Future studies, likely using qualitative methods, could explore the specific linguistic backgrounds and sectoral employment patterns of migrants from different sub-Saharan countries to better understand why language seems to matter in this case. Scholars could inquire, for example, how Moroccans’ threat perceptions may vary between migrants from Senegal or Mali (Francophone sub-Saharan countries) compared with those from non-Francophone countries.
Another future area of research concerns changes in labor market threat perceptions over time. How have citizens’ views changed before and after the “softening” of migration policy, spearheaded by Morocco’s king, who was the “guardian of the reform’s liberal spirit” (Natter 2022: 112)? Our survey took place after the reforms, so we cannot know or speculate. But, future studies—likely, again, using qualitative methods—could investigate this topic using comparative historical analysis over time. One might hypothesize that regime loyalists, taking cues from their king, would be most likely to soften their views following the policy change. Such lines of research could generate a better understanding of how labor market fears emerge, how they are reproduced over time, and whether they reflect substantive realities of the labor market.
Finally, we report mixed results on whether living in areas with high sub-Saharan migrant presence or a sudden upsurge in foreigners increases Moroccans’ labor-market threat perceptions. Indeed, it will be imperative for future work on migrants’ perceived role in the labor market to try to disentangle how actual labor market encounters and competition can be disaggregated from persistent racist stereotypes (Menin 2020; Gazzotti 2021). A “culture of silence” characterizes Morocco’s racially stratified society, and Moroccans neglect “to engage in discussion on slavery and racial attitudes” (El Hamel 2002: 29). Nonetheless, even under a culture of silence, racialized perspectives clearly shape domestic debates in Morocco, structuring opinions about migration and threats in the labor market. What is more, government policies have served to reinforce Moroccans’ latent racism against Blacks and, as Tyszler (2021: 955) avers, cultivated a “racialised category of undesirables” that associate “black skin color with illegality.” According to King (2021: 8), white Moroccans have only taken “tentative steps” toward recognizing their society’s “legacies of racial slavery and oppression.” Additional inquiry is therefore required to fully grasp how racism and the historical legacies of racist practices of social organization in Morocco may contribute to the formation of individual attitudes about migration and the labor market.
Supplemental Material
sj-pdf-1-ilr-10.1177_00197939231172481 – Supplemental material for Surveying the Landscape of Labor Market Threat Perceptions from Migration: Evidence from Attitudes toward Sub-Saharan African Migrants in Morocco
Supplemental material, sj-pdf-1-ilr-10.1177_00197939231172481 for Surveying the Landscape of Labor Market Threat Perceptions from Migration: Evidence from Attitudes toward Sub-Saharan African Migrants in Morocco by Matt Buehler, Kristin E. Fabbe and Eleni Kyrkopoulou in ILR Review
Footnotes
Acknowledgements
We also thank the Howard H. Baker Jr. Center for Public Policy and the University of Tennessee for financing a pilot study. We appreciate feedback from Rich Nielsen, Mark Tessler, Natasha Iskander, the Project on Middle East Political Science (POMEPS) labor workshop, the American Political Science Association (APSA) 2022, the International Studies Association (ISA) 2022, and the Middle East Studies Association (MESA) 2022.
This article is part of an ILR Review Special Issue on Labor in the Middle East and North Africa: Precarity, Inequality, and Migration.
The authors recognize the financial support of the Middle East Initiative and Emirates Leadership Initiative at the Harvard Kennedy School.
For information regarding the data and/or computer programs used for this study, please address correspondence to
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
