WAR, CONFLICT AND MIGRATION

The 1951 Convention Relating to the Status of Refugees defines a refugee as someone who is unable or unwilling to return to their country of origin owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion. To be granted refugee status, an asylum seeker must, therefore, have a genuine fear of being persecuted on one or more of those grounds. In the interests of brevity, the term “refugee” is also used in this chapter to refer to asylum seekers (individuals who have sought international protection, but whose applications for refugee status have not yet been assessed), as well as other internationally displaced people.

Almost two-thirds of the world’s refugees come from Syria, Ukraine, the West Bank and Gaza, Venezuela or Afghanistan. What is more, nearly half of all refugees are children, whereas children account for only a third of the world’s population. Low and middle-income countries host the vast majority of the world’s refugees. Many economies in the EBRD regions have become major destinations for refugees owing to their proximity to areas affected by conflict. Indeed, countries in the EBRD regions host 33 per cent of the world’s internationally displaced persons. Refugees tend to be younger and better educated than the average person in their country of origin, and well-designed integration policies can help them to make meaningful contributions to the economies of their host countries. On average, the integration policies of EBRD economies in the EU were marginally to moderately supportive in 2021.

At the time of writing (August 2022), at least 12.9 million people have fled their homes on account of Russia’s invasion of Ukraine.2 More than 6.3 million have left for other countries (albeit several hundred thousand have now returned to Ukraine after originally leaving the country), while there are an estimated 6.6 million internally displaced people (IDPs) inside Ukraine. In contrast with the large wave of refugees seen in 2015-16, the EU is allowing Ukrainians to live and work in its 27 member states for up to three years. Ukrainian refugees are also able to access social welfare payments, social housing, healthcare and schools.

A survey of Ukrainian refugees in Europe indicates that nearly 30 per cent are employed in their current country of residence, while 20 per cent are continuing to work remotely in Ukraine. Overall, their living conditions and the help provided by locals are both considered to be very good on average. At the same time, some respondents report having had trouble accessing financial assistance, healthcare and legal advice. Feelings of homesickness and helplessness are also common.

Ukrainian refugees are more likely to come to places with higher tax revenue per capita and areas that already had large Ukrainian communities before the war. Attitudes towards refugees have improved, with the prevalence of positive attitudes increasing by 19 percentage points in Poland relative to 2021. Almost half of all school-aged Ukrainian refugees are enrolled at local schools, while many others are continuing their education in Ukraine on a remote basis. As with other Ukrainian migrants before the war, female refugees generally work in manufacturing, support services and the accommodation sector.

This chapter starts with a global snapshot of forced migration. It then examines the profile of Ukrainian refugees in Europe, looking at their intended destination countries, their intentions to return and the challenges that they face. The last section looks specifically at Ukrainian refugees in Poland, examining their socio-demographic characteristics and their socio-economic integration to date.

It is worth noting that, relative to the size of the local population, the three countries that host the most internationally displaced persons are all in the EBRD regions: Jordan, Lebanon and Türkiye (see Chart 2.4). Indeed, countries in the EBRD regions host a third of the world’s internationally displaced persons (13.8 million when Ukrainian refugees are included). More generally, low-income countries host 15 per cent of the world’s internationally displaced persons, with middle-income countries hosting another 60 per cent, while high-income countries host the remaining 25 per cent. Box 2.1 discusses the ways in which those refugees may benefit their host countries’ economies.

Of the 9.6 million people who have crossed the border into neighbouring countries since 24 February 2022, a total of 3.3 million have stayed in those neighbouring countries (with 1.2 million currently living in Poland). In addition, 2.7 million Ukrainians are currently living in other European countries – mainly Germany, the Czech Republic, Italy, Türkiye and Spain. A substantial number have also moved back to Ukraine. In order to account for both short- and medium-term pressures on host countries’ social security systems and their societies as a whole, the analysis in this chapter uses both the number of border crossings since 24 February 2022 (which is only available for countries that share a border with Ukraine) and the number of refugees present in each country on 19 July 2022.

The results reveal that people have become much more tolerant of refugees in their countries since 2021, with the majority of respondents supporting people’s right to seek refuge in another country (see Chart 2.6). In Poland, for instance, 85 per cent of respondents agreed with that statement in 2022, up from 66 per cent in 2021, while Hungary saw an increase from 63 to 71 per cent over the same period. In Türkiye, on the other hand, the proportion of respondents with a favourable opinion of refugees declined from 70 to 66 per cent. On the whole, the war in Ukraine appears to have increased people’s openness to refugees fleeing war or oppression.

This section analyses the relative importance of those factors using data on 155 countries of origin and 138 destination countries over the period 1962-2014. In particular, the analysis relates the number of refugees from a given country of origin in a given destination country in a given year to various country-level characteristics (such as population size and income per capita), as well as measures of the linguistic, cultural and geographical proximity of the relevant pair of economies.

Refugees tend to come from poorer countries, and they tend to settle in poorer countries as well (see Chart 2.7). Distance also matters: a 1 standard deviation increase in the distance between a country of origin and a potential destination country leads to a 0.5 per cent decrease in the number of refugees in that destination country; thus, refugees are more likely to settle in a neighbouring country (that is to say, one that shares a border with the country they have just fled). Refugees are also more likely to head for countries that are similar to their own in terms of culture and the language(s) spoken. Partly as a reflection of these patterns, refugees are also more likely to end up in economies with weaker democratic institutions (as measured by their Polity score). These patterns are broadly comparable to those observed for economic migrants, with income differentials, migrant networks and demographic factors all found to be robust predictors of migrant flows.11 See also Box 2.2 for a discussion of refugees’ trust in the political institutions of their host countries.

The analysis in this section is based on an online survey of Ukrainian refugees that was conducted by Kantar, a survey company, between 14 June and 8 July 2022. That survey, which covered more than 2,600 individuals across 27 economies in Europe, included questions about their background, their reasons for leaving Ukraine, their current labour market status, their intended destination country and the hardships they had faced in their current country of residence. Respondents were reached via “social media sampling” – that is to say, via adverts on social media platforms. In line with the overall profile of Ukrainian refugees, respondents were overwhelmingly female (82 per cent) and more than half (53 per cent) were married. The largest percentage (35 per cent) were located in Poland, followed by Germany, Spain, the Czech Republic, France and Italy. Around half of all respondents had been in employment before leaving the country, with 65 per cent educated to tertiary level.

Available evidence on Ukrainian refugees’ level of education shows not only that they are more likely to be educated to tertiary level than other refugee groups, but also that they are more highly educated than the general Ukrainian population. Indeed, 2020 data for Ukraine’s working-age population showed that 56 per cent of women and 43 per cent of men were educated to tertiary level. This has been backed up by separate studies in individual countries: a survey of Ukrainian refugees carried out in Germany by the Federal Ministry of the Interior and Community found that 73 per cent of all adult respondents had a tertiary qualification, while an equivalent survey in Spain reported a figure of 61 per cent.12

According to survey data, by far the biggest reason for choosing one country over another was the presence of a pre-existing social network – that is to say, family members or friends – in the relevant destination country (see Chart 2.8). This is supported by 2020 data showing that over a million Ukrainian citizens were resident in EU countries before the war, mostly in Italy (236,000), the Czech Republic (163,000), Poland (145,000), Germany (135,000) and Spain (105,000).13 Prior to the war, Ukrainian citizens accounted for large percentages of total immigration in four EBRD economies: Poland (45 per cent), Lithuania (34 per cent), the Czech Republic (26 per cent) and Hungary (14 per cent).

Geographical proximity to Ukraine and the likely availability of work also played an important role in refugees’ decision-making, as did cultural proximity and schooling options for children. These findings are broadly consistent with the evidence discussed earlier in the chapter in relation to a larger sample of countries.

Most respondents intended to stay in their current country of residence for the time being, with only 3 per cent planning to move on to another country (outside Ukraine) in the near future (see Chart 2.9). Among those who reported a desire to move on, the top five preferred destinations were Canada, Germany, Poland, Belgium and the United Kingdom. In the majority of cases, these survey responses may only indicate short-term plans, with 64 per cent of respondents reporting that they intended to go back to Ukraine soon or when it felt safe to return.

In part, this reflects the extensive support that host countries have given refugees in other areas (such as employment and benefits), with nearly 30 per cent of all respondents already working in their host country at the time of the survey, despite the language barriers. In addition, one in five people reported working remotely in Ukraine while living in their host country. Respondents reported that living expenses were being financed by a combination of financial support from host countries, family savings and income from work.

In terms of the conditions and public services in their host countries, refugees are mostly of the view that the help and support provided by locals, their general living conditions, their access to education and their housing conditions are all very good (see Chart 2.11). The most problematic aspects are access to legal advice, medical care and financial assistance, with less than one-third of respondents reporting a lack of satisfaction in those areas.

Overall, these findings highlight Ukrainian refugees’ appreciation for the significant and timely support that European citizens and governments have given them. In contrast with the large wave of arrivals seen in 2015-16, the EU is allowing Ukrainians to live and work in its 27 member states for up to three years. Ukrainian refugees are also able to access social welfare payments, social housing, healthcare and schools.

Thus, refugees are often the most vulnerable group of immigrants and face steep barriers to economic and social integration.15 This makes it even more important that host countries have effective integration policies to enable refugees to participate fully in the local economy.

In order to gain greater insight into variation in integration policies for refugees across countries, the analysis in this next section looks at refugee integration scores compiled by the National Integration Evaluation Mechanism (NIEM) for 14 EU member states on the basis of policies in place in 2021.16 Scores range from 0 (denoting the least advantageous policies) to 100 (indicating the best policies), evaluating integration in five key areas: housing, employment, healthcare, education and adult language courses.

On average, the integration policies of EBRD economies in the EU are marginally to moderately supportive (that is to say, they have integration scores of between 25 and 75; see Chart 2.12). Those scores range from 33 in Hungary to 63 in Lithuania – below the levels observed in Sweden (73) and France (68). To some extent, this gap between the EBRD regions and developed economies reflects the fact that high-income European countries have been receiving large numbers of refugees for many years, resulting in more established asylum policies and integration frameworks. Over the last 10 years, however, the asylum and integration frameworks of those EBRD economies have become more comprehensive, reflecting the harmonisation of laws and regulations following their accession to the EU.17

At the same time, there is considerable variation both across and within countries when it comes to the different policy areas. The widest gap can be observed for language learning, with 93 points separating the countries with the highest and lowest scores. Language skills and knowledge about the host country’s institutions and social norms are key to refugees’ ability to integrate. While NIEM reports good provisions in some EBRD economies, there are few – if any – publicly funded language and social orientation courses in Greece, Bulgaria and Hungary. Such courses need to be made available for an extended period of time and tailored to the requirements of specific groups (such as unaccompanied minors), as well as individuals with differing levels of proficiency. Countries such as the Czech Republic, Latvia and Romania score highly by providing courses without additional costs or compulsory attendance. In addition, needs-based translation and interpreting assistance is also beneficial to refugees.

Housing is another key area where certain countries – particularly EBRD economies – fail to establish sufficiently supportive policies. In all countries except Greece and Slovenia, refugees have a legal guarantee promising them access to housing on the same terms as locals. However, a lack of government support, high rental costs in the free housing market relative to refugees’ disposable income, the language barrier and discrimination often make it difficult for refugees to find suitable accommodation. Sweden and France have the best housing conditions, with their supportive housing policies including government schemes to identify suitable housing, coordination between public and private actors, and the provision of housing benefits and targeted information for refugees. EBRD economies could improve by reducing administrative barriers (such as the requirement to leave shelters shortly after being granted protection and the need for specific documents) and providing financial and organisational support when refugees come to find accommodation of their own.

When it comes to healthcare, most countries have legislation ensuring that refugees have access to appropriate facilities. Most countries also have fairly favourable legal conditions facilitating refugees’ access to the labour market.

The legal provisions of most host countries are favourable towards Ukrainian refugees in terms of access to housing, healthcare, education and so on, given the EU’s activation of the Temporary Protection Directive. A major challenge for EBRD economies will be following through with supportive measures that ensure not only protection but also adequate living conditions and opportunities for integration.

More than 80 per cent of all studies focusing on these effects find that refugees have a positive or neutral impact on the labour market outcomes of local workers, while fewer than 20 per cent report negative and statistically significant impacts. This is mainly because refugees and local workers typically have different sets of skills and compete for different types of job.

Using data from a follow-up to the Enterprise Survey conducted by the EBRD, the European Investment Bank (EIB) and the World Bank in 2018-20, Chart 2.13 examines firms’ views regarding the contributions that refugees make to the labour market. That follow-up survey covered a total of 815 firms across 15 economies in the EBRD regions (see Box 3.3). In particular, it asked firms about the hiring of refugees and the impact that refugees could have on their access to labour, as well as customer outreach.

On average, firms tended to view refugees’ labour market participation positively (see Chart 2.13). Around 40 per cent of all respondents reported that refugees were likely to have a positive effect on their firm’s ability to access skilled labour, compared with 10 per cent of firms expecting a negative effect (with half of all firms having a neutral view). Similar figures were observed when firms were asked about access to cheap labour. Firms that would consider hiring refugees were more likely to believe that an influx of refugees would have a positive impact on their access to cheap and skilled labour.

Similar to other EU member states, Poland has, in line with the EU’s Temporary Protection Directive, given Ukrainian refugees access to social services and allowed them to live, work and study in the country for up to three years without having to apply for asylum. Refugees who struggle to find a job are entitled to register with district employment agencies to receive assistance and professional guidance.

Having the right to start working as soon as they arrive in Poland improves refugees’ long-term employment prospects.20 Research shows that refugees who face an employment ban on arrival in the host country (pending a review of their case) are 15 per cent less likely to be employed in subsequent years, even after the necessary authorisation has been granted (as those refugees tend to opt out of the labour force).21

By 30 June 2022, around 32 per cent of all female Ukrainian refugees were in employment locally (see Chart 2.15). The employment rate for Ukrainian men was higher (at 61 per cent), but those working men were a much smaller group (around 65,000) in absolute terms. What is more, 6 out of 10 women had travelled to Poland with children, thus limiting their ability to take on employment. Employment rates for both men and women have risen steadily over this relatively short period, attesting to the good progress that has been made in terms of refugees’ integration into the labour market. Consequently, remittance flows from Ukrainians in Poland are expected to increase sharply as refugees seek to support friends and family at home.22

Most female Ukrainian refugees are working in manufacturing, support services or the accommodation sector (see Chart 2.16). This pattern is very similar to that observed for female Ukrainian migrants who came to Poland before the war, which partly reflects the existing labour market shortages in Poland. The similarity between those two employment profiles also highlights the role of information-sharing among migrants through pre-existing social networks.23

Chart 2.17 reveals a number of patterns. Refugees tend to settle in counties with higher income per capita (captured here by personal income tax revenue per capita). This may reflect better employment prospects in richer counties, as well as a superior capacity to provide assistance to refugees (in terms of social care, housing and education, for example). Indeed, administrative capacity and income per capita tend to be strongly correlated both across and within economies.

Ukrainian refugees are also more likely to settle in counties with larger numbers of pre-war migrants from Ukraine, which highlights the role of social networks. The relative importance of such networks is even larger when it comes to successful integration into the labour market: a 1 standard deviation increase in the number of pre-war Ukrainian migrants as a percentage of the population is associated with a 0.7 standard deviation increase in refugees’ employment rate, taking into account other factors such as average local income per capita.

While refugees are more likely to reside in counties with lower urbanisation rates (perhaps because accommodation is more readily available), employment rates are higher for refugees residing in more urbanised locations, perhaps reflecting tighter labour markets in metropolitan areas.

Only the urbanisation rate is strongly correlated with the school enrolment rate for refugee children (with enrolment rates being higher in urban areas). This may, in part, reflect the fact that many children have been able to continue their Ukrainian schooling online, at least for the remainder of the 2021-22 academic year.

A recent study focusing on the United States of America found that refugees were, on average, more entrepreneurial than both the native population and people who had migrated to the country for economic reasons.28 Indeed, refugees have been responsible for some major innovations over the years, with examples including Sergey Brin, the co-founder of Google, Jewish German scientists (such as Albert Einstein) who revolutionised science in the United States of America and, a few centuries back, the Huguenot refugees from France who brought textile manufacturing technology to Germany and London.29 Such entrepreneurial behaviour may reflect the considerable tolerance of risk that is shown by refugees when they embark on long, perilous journeys fleeing conflict.

Refugees’ contributions to their new host countries do not always come automatically. The right policies need to be in place to allow refugees to make meaningful contributions to the economic dynamism of their local areas. Evidence from 40 years of refugee policies in Denmark highlights the importance of employment support and language training.37 Access to the labour market (and the structure and future prospects that it provides) also reduces problems relating to violence among asylum seekers.38

In many instances, however, refugees are not granted the right to work, either out of concern that they will compete with domestic workers for jobs or because they are not expected to stay for long. In Türkiye, for example, a lack of work permits drove millions of Syrians into the informal sector; in Colombia, millions of Venezuelan refugees were given visas that were valid for just two years, which made it very difficult for many of them to find jobs; and, in Bangladesh, over a million Rohingya refugees have been confined to camps without the right to work.39

Refugees’ prior exposure to political institutions will differ from that of the local population in their host country. For instance, refugees who have escaped a country with a high level of corruption may potentially expect such corruption to be pervasive, leading them to place less trust in the political institutions of their host country.42 To investigate that relationship, this box uses data from eight waves of the European Social Survey (which were conducted at two-year intervals over the period 2004-18 in 38 European countries), together with country-level measures of corruption (executive bribery and embezzlement, judicial corruption, legislative corruption, and public-sector bribery and embezzlement) as captured by the Varieties of Democracy (V-Dem) dataset.

This analysis compares refugees who have come from the same country of origin and are resident in the same host country, but have been exposed to differing levels of corruption in their home country as a result of leaving the country at different points in time. Specifications also take account of the year in which the survey was conducted, the refugee’s year of birth, demographic and labour market characteristics, and country-year fixed effects for both the country of origin and the host country. This allows us to net out any impact of time-invariant unobservable characteristics of current and former countries of residence. The results are presented in Table 2.2.1.

In fact, exposure to higher levels of corruption in the country of origin has a positive impact – rather than a negative one – on trust in the political institutions of the host country, with exposure to corruption during an individual’s most impressionable years (between the ages of 18 and 25) appearing to matter most in this regard. An individual with a high level of exposure – 7.67 on a scale ranging from 0 (no corruption) to 10 (large-scale corruption) – is estimated, on average, to be 5.4 percentage points more likely to trust the institutions of his/her host country than an individual with typical exposure to corruption. Given that the average level of trust is 60 per cent, this effect is sizeable.

These findings are in line with the results of a study by Kahneman and Tversky (1979), who argued that when current institutions compare favourably with the institutions that attitudes and expectations have been based on, they are viewed more positively.

These findings go some way towards alleviating the often-stated concerns about refugees’ lack of adequate regard for host countries’ institutions, values and cultures, or refugees’ corrosive influence on political trust across society as a whole in host countries.

Comprehensive statistics on claim-processing times are not available. Data for Sweden and Germany point to substantial heterogeneity across countries of origin, similar to the patterns observed for acceptance rates. In 2021, for instance, it took Germany an average of 1.6 months to process asylum claims made by citizens of Bosnia and Herzegovina, compared with 14 months for Somali citizens.46 This pattern may reflect differences in the difficulty of verifying grounds for asylum, as well as varying volumes of asylum applications. It may also be affected by public attitudes towards migrants and differences between applicable policy frameworks.47

However, favourable asylum policies (such as allowing asylum seekers to work and arrange their own accommodation, and making language training available) may mitigate the detrimental impact of longer waiting times and support refugees’ accumulation of human capital. In Sweden, which has such policies, a recent study found that longer waiting times had no effect on refugees’ mental health and a positive impact on their entry into the labour market.51

Under a special legislative provision enacted on 12 March 2022, Ukrainian refugees can register for a Polish national identification number, which entitles them to access public healthcare and education, as well as social welfare payments (including child allowances). Data on the number of registered refugees, as well as their age and gender, are based on the PESEL registry and are available at county level (equivalent to NUTS 4 – level 4 of Eurostat’s Nomenclature of Territorial Units for Statistics). Ukrainian refugees are allowed to work without registering for a PESEL number, but their employers are legally obliged to notify the local labour office of such employment. Data on refugees’ work permits have been obtained from the Polish Ministry of Family and Social Policy. Figures relate to the end of the month and are available for the period from March to June 2022. Additional data cover the number of refugee children enrolled at Polish schools, broken down by county.

These data are complemented by county-level information on Ukrainian citizens who have been working in Poland since 2019, which includes details of their type of residence permit and contract, their gender, the industry they are employed in and their occupation (at the two-digit level of the International Standard Classification of Occupations). County-level characteristics such as municipal spending are taken from Statistics Poland. All data have been aggregated at the level of the 340 counties defined by the Ministry of Family and Social Policy.

The outcome variables are: (i) the popularity of a destination among refugees, calculated as the number of refugees in a county relative to the population of that county; (ii) the employment rate for refugees, calculated as the number of active work permits relative to the number of refugees between the ages of 15 and 64 who are registered in a county; and (iii) the school enrolment rate for refugees, calculated as the number of refugee children who are enrolled at primary school relative to the number of refugees between the ages of 7 and 13.

A key variable of interest is the number of Ukrainian migrants who were working in Poland before the war (as at 23 February 2022) relative to the population of the county in question, which captures the role that social networks play in migration decisions. Other variables include the urbanisation rate, as well as the ratio of teachers to students and municipal spending per capita (as measures of the quality of public goods). Local economic conditions are captured by the pre-war unemployment rate (as at January 2022), personal income tax revenue per capita (as a proxy for income per capita in the local area) and the number of hotel beds per capita.

Jordan’s response to this influx of refugees has leveraged international support, focusing on adequate provision of municipal infrastructure, as well as programmes fostering skills and integration into the labour market. Jordan’s Ministry of Planning and International Cooperation has adopted a centralised and transparent approach to managing support for Syrian refugees, taking account of the concerns of host communities and the need for investment in local infrastructure. The country’s refugee response plan (which has been implemented in cooperation with the EBRD) has focused on investment in the management of solid waste, transport links, renewable energy and water supply. This has helped to preserve social cohesion, despite significant increases in the populations of certain municipalities.

When it comes to the development of skills, coordination appears to have been weaker, with organisations running similar programmes in competition with each other and insufficient attention being paid to leveraging synergies between programmes and tailoring them to the needs of the labour market.

Flexible work permits have enabled refugees to move between jobs within a given sector, while short-term work permits targeting seasonal occupations have also supported job creation. However, with Jordan’s unemployment rate reaching 23 per cent in 2022, many refugees with work permits are struggling to find jobs. The Covid-19 pandemic has driven many refugees into poverty, with food insecurity on the rise, exacerbated by high food and energy prices.

Balancing the interests of nationals and the needs of refugees remains a challenge. In response to rising unemployment on account of the pandemic, the Jordanian authorities have restricted non-nationals’ access to jobs in some sectors. This has hampered the competitiveness of certain employers and made it harder for some refugees to find employment that matches their skill-sets. Broader involvement of employers in policy consultations can help to refine the assessment of skill gaps and limit negative consequences for sectors with limited availability of skilled nationals. At the same time, stricter enforcement of minimum wage and labour protection laws for all workers could help to alleviate nationals’ concerns about unfair competition and improve working conditions for refugees.

Financial and non-financial support for self-employed workers has remained limited. This has been a particular challenge for women, who shoulder a disproportionate share of the burden when it comes to childcare and looking after elderly relatives. In order to assist female workers, the Microfund for Women (MFW) has begun offering loans to Syrian women who are reliant on Jordanian family ties or guarantors for security. Building on the resulting credit history and market experience, the MFW has gradually rolled out group loans and individual “Tatweer” (development) loans to Syrian refugees, as well as providing non financial assistance aimed at improving their business management and digital skills.

Registered Syrian refugees of school age have free access to Jordanian state schools, with additional costs being covered by the international community. As a result, around 136,000 Syrian children are currently enrolled in formal basic education in Jordan. More than 200 schools in host communities are continuing to operate a two-shift system, with afternoon shifts earmarked for Syrian children. Meanwhile, the private Luminus Education initiative has secured over US$ 19 million in grants from the Jordanian government and international donors to help refugees to complete tertiary education and enter the labour market. Those multi-year grants have covered the tuition fees of more than 3,000 students. Luminus also works closely with potential employers to identify jobs that are available to refugees when their studies end.