Forced migration: evidence and policy challenges

Abstract

This paper presents a summary assessment of this issue of the Oxford Review of Economic Policy, on forced migration. The issue is concerned with four important questions: (i) What are the general mechanisms by which forced migrants should be managed, and what frameworks should be used for supporting them? (ii) How can policy help refugees integrate into host economies; and what are the likely consequences of this integration? (iii) How are host communities likely to respond to the influx of refugees, and how can policy help to smooth this transition? and (iv) What role can policy play to encourage resilience among refugees and internally displaced people—and, one day, potentially support their return? Drawing from a diverse set of experiences and country case studies, the invited authors—who range from academics to policy practitioners—present and discuss current evidence and draw from their expertise to offer insights on these general themes in the economic policy response to forced migration. Among others, some of the recurring ideas for the design of policy include the need of anticipatory, systematic, and long- term approaches to the ‘management’ of forced displacement; the importance of building evidence, quantifying impacts, and understanding the distributional consequences of forced migration; and finally, the importance of bridging a gap in how the evidence is communicated and understood in the broader community.

Introduction

Forced migrants are those who leave their home, their region, or their country of birth, as a result of persecution, conflict, violence, human rights violations, and other events that seriously disturb public order (UNHCR, 2021). Since the 1970s, there have been at least 15 conflicts that have forced at least one million people or more to leave their country of birth—and many other countless clashes that have displaced large numbers of people within their own national borders (so-called ‘internally displaced people’—IDPs).

The most recent of these tragic conflicts is, of course, rarely far from our screens or from our minds. On 24 February 2022 (this year), Russia invaded Ukraine. This invasion—which followed the 2014 annexation of Crimea, and the subsequent occupation of the Ukrainian oblasts of Donetsk and Luhansk—clearly poses new and profound challenges for global cooperation under the international system. It is already clear that one of the most consequential policy issues concerns the treatment of Ukrainian refugees. Over the opening weeks of the Russian invasion, approximately 130,000 Ukrainians fled their country every day: a total of about four million refugees over the course of the first month alone (UNHCR, 2022a). This situation added to the difficulties already faced by large number of IDPs and other displaced people in Ukraine since 2014. Indeed, by 2016 there were already close to two million IDPs in addition to over thirty thousand people seeking asylum¹ (UNHCR Statistics, 2022). The plight of these people demands compassionate and thoughtful responses from governments around the world—both at the national and local levels, and particularly in Europe.

As early as 1951, the parties to the Convention Relating to the Status of Refugees agreed a single definition in international law for the concept of a refugee,² and committed to a set of core obligations concerning the treatment of refugees—including, in particular, that refugees have a basic entitlement to protection and support.³ By mid-2021, there were close to 25 million refugees globally, as recorded by the United Nations High Commissioner for Refugees (UNHCR, 2022b). As Figure 1 shows, the seven decades since 1951 witnessed a substantial increase in the total number of refugees worldwide. Moreover, the nature of the refugee experience has changed substantially, too; this is reflective, in particular, of changes in the nature of conflict, changes in the methods for refugee protection, and important shifts in the global political structure. For instance, the 2015 Syrian and European refugee ‘crises’ were the driving force behind the 2018 Global Compact on Refugees (Betts, 2018), but at the same time also had a profound and divisive impact on public opinion and voting behaviour in many high-income destination countries (Dustmann et al., 2019; Dinas et al., 2020; Steinmayr, 2021). However, it is countries in the Global South—in particular, those bordering major conflicts—that host the majority of refugees. Displacement has become an urban issue and refugee camps are becoming less common (Vos and Dempster, 2021; Crawford, 2021). There is an increasing number of protracted displacements;⁴ in turn, this keeps millions of refugees in legal limbo for decades. Finally, as shown in Figure 1, the number of people displaced within their own countries—50.9 million by mid-2021—is much larger than the number of those displaced internationally, and these numbers continue to increase.⁵

Figure 1:

Number of refugees under UNCHR Mandate (1950–2021).Note: End year stock population totals. The number of refugees include displaced Venezuelans but do not include Palestine refugees under the United Nations Relief and Works Agency (UNRWA) Mandate. As of mid-2021, the number of Palestinian refugees was estimated at 5.7 million. IDPs are those reported in UNHCR Statistics under the UNHCR population of concern (UNHCR, 2022b).

Source:  UNHCR Statistics (2022).

In this context, policy-makers face a set of related questions. First, what are the general mechanisms by which these refugees should be managed, and what frameworks should be used for supporting them? Second, how can policy help these refugees to integrate into host economies—both at the macro and the micro level—and what are the likely consequences of this integration? Third, how are host communities likely to respond to the influx of refugees, and how can policy help to smooth this transition? Fourth, what role can policy play to encourage resilience among refugees and internally displaced people—and, one day, potentially support their return?

These are the core questions that are tackled by this issue of the Oxford Review of Economic Policy. Specifically, the issue draws together a set of thinkers with particular expertise on forced migration. We invited the authors in May 2020 and most of the papers were received in final form just prior to the Russian invasion of Ukraine. In this sense, the journal issue speaks to general themes in the economic policy response to forced migration; indeed, together, the papers draw on experiences from a very diverse set of countries. These include Syria, which currently has the largest population of refugees abroad (6.7 million), and Colombia—a country with a long history of conflict that, with 8.3 million, has the largest number of internally displaced people (UNHCR, 2022b).⁶ The articles also draw on the experiences of important host destinations, including Jordan, Lebanon, and Turkey—the three main destinations of Syrian refugees, which, as of mid-2021, were together hosting over five million Syrian refugees. Adding the perspective of high-income countries, the issue draws on experiences from Denmark, Sweden, and the United States. The experience in these countries is different in at least three distinctive ways: first, refugee populations in these countries are small relative to the host, both in population terms and in fiscal terms. Second, and as a result, the policy focus has been about integration, dispersion, and assimilation and less on ‘emergency’ management and hosting. The third is that in the case of high-income countries we typically have much richer cross-sectional and time-series data. Therefore, in this case not only are the host-country policy considerations different, but the capacity to inform these through research and evidence is greater. Finally, the issue also draws attention to challenges brought by the return of refugees and the impacts on different aspects of social cohesion, by looking at the Great Lakes refugee crisis and the return of Burundians refugees from Tanzania.

The journal issue is organized in four separate parts—each relating to one of the questions that we posed earlier. Part I discusses mechanisms for managing refugees—‘mechanisms’ both in the general sense of the international institutions and processes that manage the response to refugee crises, and in the specific sense of using insights of modern market design techniques to match refugees effectively with local services. Parts II and III consider integration of refugees into host communities. Part II focuses on refugees and the macroeconomy, with Part III taking the microeconomic impacts through local labour markets and host communities. Part IV then discusses the longer-term path for refugees and internally displaced people: it considers issues of recovery, resilience, and return. The issue concludes with a postscript: a summary discussion of the likely policy implications for supporting refugees from the ongoing crisis in Ukraine.

Part I: Refugees and international mechanisms

Part I of the issue considers the particular systems that are used to respond to refugee crises. Some important questions here are, for example, what mechanisms should be used to deliver aid and public services to those that most need it? What can be done by the international community to help and assist host countries receiving refugees? What sort of global financing mechanisms should be in place in order to support the different responses to refugee crises? The first paper in the issue—by Grant Gordon and Ravi Gurumurthy (2022)—provides a framework to think about these questions and it poses a vision of how the humanitarian sector should look for the next 10–20 years. The paper takes stock of the evolution in the responses to forced displacement, identifies notable innovations, and proposes a way forward with an innovation agenda focused on delivering cash transfers and digital aid. It also highlights the importance of ‘compacts’—at the country level and with the support of the international community—that expand entitlements to work, and access to education and public services. An important aspect of this vision and of the future of innovation in the humanitarian sector, as noted by the authors, is that it should be grounded in pre-positioned—anticipatory—financing and policy. In this, the authors draw parallels with the ideas of Daniel Clarke and Stefan Dercon on the importance of a long-term approach to the management of crises. The authors also highlight important ethical issues that need to be considered as innovations in the sector take place.

Justin Hadad and Alexander Teytelboym (2022) then focus more specifically on mechanisms for improving the management of refugee settlement; they do so through the lens of market design. Hadad and Teytelboym start with a fundamental concern: the current refugee resettlement system is inefficient, in the sense that there are too few resettlement places and, when refugees are resettled, they often go to locations where they might not thrive. From this starting point, the paper then highlights several ways in which the market design paradigm can help: better matching between locations and refugees (therefore improving the prospects of success in outcomes for the refugee and the host), and creating incentives to increase the participation of countries in resettlement schemes. Importantly, market design can mitigate some of the worst inefficiencies and unfairness in the current system—improving on the status quo by incorporating refugees’ preferences, communities’ priorities, and economic outcomes. There is substantial scope for market design methods to improve practices at local resettlement agencies and at the international level. However, ultimately it is only political will that can increase resettlement.

Part II: Refugees and the macroeconomy

Michael Clemens (2022) begins Part II of the issue, with a critical review of the literature on the consequences of refugee arrivals on the national economy. Clemens acknowledges immediately in his paper that economic gain is not the purpose of refugee and asylum policy. Nonetheless, it is crucially important—and, indeed, timely—to quantify the likely magnitude of such economic impacts, particularly given the tone of so much of the political discourse concerning the economic costs and benefits of accepting refugees and asylum-seekers. The paper does exactly that: it measures the consequences to the US economy of the decision to reduce refugee arrivals from 2017 to 2020. The estimates are substantial, and negative: the paper suggests that, on a conservative estimate, a net loss to government revenue of almost US$7,000 per year per ‘missing refugee’, and an overall cost to the economy of about US$31,000 per year per ‘missing refugee’. As Clemens explains, these figures ‘are large in one sense, small in another’; in particular, the costs are minimal relative to the overall size of the US economy.

Trade and foreign direct investment (FDI) together form one very important angle from which refugees might impact a host economy. This is the theme of the paper by Dany Bahar, Christopher Parsons and Pierre-Louis Vézina (2022) who document a ‘diversity dividend’ and a ‘dynamism dividend’ that may allow refugees to be particularly well placed to develop new opportunities for trade and cross-border investment. Among many others, an example is the increased trade between Vietnam and US states hosting Vietnamese refugees, and the development of products such as the Sriracha chilli sauce, a well-known and globally recognized staple condiment.⁷ Importantly, another aspect of these cross-border interactions is the role played by refugee diasporas in helping processes of post-conflict reconstruction. As pointed out by the authors, diasporas maintain strong networks with their countries of origin and are well placed to lead and encourage the funding of needed resources to help the development of their country of origin. The authors illustrate with examples of developmental actors and projects that aim at finding ways to channel these efforts to foster peace and development (e.g. The DIASPEACE project). The authors then outline a set of potential policies to encourage refugees in their support for reconstruction and economic development in their origin countries. These include establishing regulations and policies to differentiate between refugees and migrants, providing labour market access to refugees, reducing the costs of remittances, finding mechanisms to leverage the role of diasporas, and—when appropriate—facilitating and incentivizing post-conflict return.

Part III: Host labour markets and host communities

In the first paper in this section, Alexander Betts and Olivier Sterck (2022) note that in low- and middle-income countries there is significant variation in policy responses towards refugees; they ask, ‘why do some states give refugees the right to work, while others do not?’ Betts and Sterck pose competing theories based on interest-based, norms-based, and identity-based accounts; the authors outline potential mechanisms through multi-level bargaining at the global, national, and local levels to explain what might determine compliance with refugee norms. To test these potential explanations, the authors use a qualitative comparative case study together with a rich quantitative dataset, finding that norms-based and multi-level bargaining explanations do indeed explain compliance with refugee norms. In particular, the authors find that the de jure right to work is associated with payoffs at the ‘national’ level (i.e. being a signatory of the 1951 Convention) whereas de facto rights are associated with payoffs at the ‘local’ level (i.e. the degree of decentralization). An important implication of this paper—as highlighted by the authors—is the importance of creating incentives both at the national and local levels in order to promote compliance of refugee norms.

Nordic countries have a wealth of high quality data that allows researchers to look at important aspects of refugee integration and the impacts in hosting communities—and two papers in this issue draw on insights from these data. First, Jacob Nielsen Arendt, Christian Dustmann and Hyejin Ku (2022) review 40 years of evidence on the impacts of different immigration and integration policies on the short- and long-term outcomes of refugees in Denmark. They focus on the Danish evidence to date of five of the most common types of post-arrival policies in high-income countries: dispersal accommodation, employment support, integration and language programmes, welfare benefits, and conditions for permanent residency. A particularly important lesson drawn from the evidence is the need to assess and recognize the potential trade-offs and unintended consequences of changes in policies. For example, while the objective of dispersal policies is to distribute the burden of hosting refugees across all hosting communities, the initial place of settlement can have both immediate and long-term impacts on the labour market performance of refugees. Equally, while employment support policies are desirable, they could crowd out enrolment in language and integration programmes which are important for long-term integration outcomes. Most compelling were their findings regarding welfare. While reductions in benefits do seem to have an initial positive response in employment, there are other consequences of these reductions in disposable income: these include higher criminal activity of refugees and their children.

Sandra Rozo and Maria Jose Urbina (2022) take advantage of the dispersal policy in Sweden to look at the impact of hosting refugees on natives’ attitudes in hosting communities. The authors find that increased shares of refugee inflows translate into lower support for immigration in the hosting communities. These attitudes are further magnified by concurrent changes in economic conditions of the host. Further, a demographic characterization indicates that those holding more negative attitudes are more likely to be young males, with less wealth, and who work in blue-collar occupations. An important implication of this study is that policies aimed at promoting social cohesion towards refugees can usefully be informed by a better understanding of who is most likely to oppose refugees. This paper nicely adds to and complements the recent literature on the impacts of refugee inflows on public opinion and voting behaviour.

While the above two papers focus on high-income countries, many host destinations are low- and middle-income countries, where informality is an important feature of the labour market. Norman Loayza, Gabriel Ulyssea, and Tomoko Utsumi (2022) use a structural spatial model that the authors had previously developed to formally test and analyse the impacts of the mass inflow of Syrian refugees in Turkey. An important finding is that low-skill workers bear the burden of the costs, as the level of informality increases and wages decline for these workers. However, an interesting implication of this model is that since tax revenues and profits per worker also increase, the losses for low-skill workers can potentially be reversed through tax redistribution and could even lead to a net gain in income per capita in most affected regions.

Part IV: Recovery, resilience, and return

The first paper in this section discusses and reviews over 20 years of research on the dynamics and consequences of forced internal displacement. The analysis is focused on Colombia—the country with the largest number of IDPs globally. As noted by Ana María Ibáñez, Andrés Moya, and Andrea Velázquez (2022), Colombian IDPs are lawfully recognized as victims of the conflict in what is perhaps one of the largest peace-building reparation programmes. The authors identify different mechanisms through which forced displacement can make IDPs vulnerable and trap them into (persistent and chronic) poverty. These include the loss of physical assets, the erosion of human capital, the loss and disruption of social networks, and psychological and behavioural impacts (loss of psychological assets and capacities). The authors further discuss the evidence on the impact of different policies to assist and support IDPs (IDP registration, anti-poverty programmes, bespoke programmes for IDP, and Reparations and Land Restitution) and identify lessons for other contexts and countries affected by forced displacement.

The paper by Sarah Stillman, Sandra Rozo, Abdulrazzak Tamim, Bailey Palmer, Emma Smith, and Edward Miguel (Stillman et al., 2022, in this issue) focuses on the socio-economic outcomes of Syrian refugees in Jordan. The paper discusses the first round of results of an important ongoing academic effort to track the outcomes of Syrian refugees. The Syrian Refugee Life Study (S-RLS), first launched in 2020, is a representative longitudinal study (2,500 households) of the socio-demographic and other characteristics of the Syrian population in Jordan. The results of the first wave of data are sobering. Syrians—perhaps not surprisingly—are more vulnerable in terms of poverty and other economic outcomes (especially those living outside camps) compared to the Jordanian population, and the impacts of the Covid-19 pandemic has likely magnified the growing gap between refugees and non-refugees.

During the lockdown, Syrian refugees had an average reduction in per adult income of 80 per cent and, after the lockdown, the number of households with positive labour income declined by 12.4 per cent. The prospects of return for the Syrian population are not very optimistic: the large majority of refugees are not hopeful that the war will be resolved any time soon and over half of them are not planning to return in the near future. Similar to what the previous paper highlighted, a common theme for refugees is the state of their mental health and higher likelihood of depression.

Complementing the S-LRS above, the next paper by Caroline Krafft, Bilal Malaeb, and Saja Alzoubi (2022) addresses the question: ‘How do policy approaches affect refugee economic outcomes?’ Focusing on education, work permits, cash assistance, welfare, food aid, and the consequences of encampment in Jordan and Lebanon, the authors discuss and assess the relatively scarce evidence on the impacts of the different policies on the refugee population (and some of the externalities on the hosting populations). While the two countries are not necessarily directly comparable, the authors highlight important commonalities and differences in terms of policy effectiveness. They also raise two important points. First, the importance of recognizing the protracted nature of the Syrian conflict. This will likely require a policy effort geared towards refugee integration and a shift to long-term development goals. A second important point is the need for better data collection—including longitudinal data—and the need for good quality impact evaluations to best inform policy-makers and other stakeholders. The previous study (i.e. the S-LRS) has taken important steps in this regard.

A final question is: what happens when refugees return? There is not much that we currently know about refugee return; from what determines the decision and timing of return to what are the consequences on the communities to which refugees return. The final paper in this section focuses on the latter; it looks at the consequences of refugee return on social cohesion. Building on their previous work⁸ on the consequences of post-conflict refugee return to Burundi after the Great Lakes conflict, Isabel Ruiz and Carlos Vargas-Silva (2022) look at indicators of support across households, violence and reconciliation, trust, and participation in community groups and the differences between stayees and returnees Their results indicate the possibility of new migration-related societal divisions in affecting post-return social cohesion. These new migration-related divisions suggest that groups which were previously coherent (e.g. based on common ethnicity) can now be split given their location during the conflict (e.g. stayees versus returnees). As noted by the authors, moving forward, policy efforts should include an understanding the dynamics behind new migration-related divisions.

Postscript: The Ukrainian crisis

As noted earlier, most of the papers in this journal issue were received in final form just prior to the Russian invasion of Ukraine. To discuss the policy implications of this ongoing war, Vlad Mykhnenko, Elliot Delahaye, and Nigel Mehdi generously agreed—on short notice—to provide an additional paper for the issue (Mykhnenko et al., 2022). Their paper begins not in late February 2022, but in late February 2014: when Russia occupied Crimea, and then moved to occupy Donetska and Lukanska. As the authors explain, these acts—and the 8 years of violence that have followed—caused massive forced migration, both to other parts of Ukraine and to other countries; it also caused over 13,000 conflict-related casualties. The key contribution of the paper is to draw upon the experiences of internally displaced Ukrainians—which the authors do using a gravity model, to characterize the relationship between individual characteristics and the migration decision—to make policy recommendations for responding to the humanitarian crisis currently unfolding. The authors highlight four key lessons: (i) heterogeneity among displaced Ukrainians is critically important in understanding their migration decisions, (ii) permanent resettlement will likely be necessary, given the protracted nature of the conflict, (iii) large-scale migration is likely to occur over long geographical distances, requiring general humanitarian support from many different state actors, and (iv) local programmes offering support, while valuable, are unlikely to have substantial impacts on migrants’ choice of destination.

The path ahead

The recent conflict in Ukraine has reminded us that conflict can escalate quickly and, with that, large population exodus ensues. This presents challenges not only for countries receiving and hosting these vulnerable populations, but it also challenges and questions the responsibilities of the international community. In this context, academic research on displaced populations has become of great importance for policy-makers and other stakeholders.⁹ The papers in this issue speak to many of these challenges and offer important evidence and possible policy solutions. They also speak directly to UNHCR’s ‘durable solutions’ to refugee situations: voluntary repatriation, resettlement, and integration within the host community, and to the stated objectives of the 2018 Global Compact on Refugees. Indeed, many of the papers raise the need of systematic and long-term approaches not only in terms of durable but also sustainable responses and solutions.

It is important, however, to recognize that while all the papers in this journal issue have focused on displacement due to war and conflict, an important challenge ahead relates to climate change and environmental displacement. Indeed, the Global Compact on Refugees has recognized this as an important root cause and one that ‘interacts with the drivers of refugee movements’ (United Nations, 2018).

Finally, many of the responses needed in forced migration settings fully depend on political will and the responses of the electorate. The salience of refugee migration in national politics has been one of the driver forces behind populist political parties and political polarization—both in high-income and middle- or low-income contexts (Altindag and Kaushal, 2021; Rozo and Vargas, 2021). This highlights the importance of quantifying the impacts of forced migration—and, equally critical, the importance of bridging a gap in how this evidence is communicated and understood in the broader community.

We thank Alison Gomm for excellent support and editorial work during the production of this issue, and to members of the journal’s Editorial Board and invited referees for very useful comments and discussions.

Footnotes

¹

Asylum-seekers are those seeking protection from persecution and serious human rights violations in another country, but are awaiting a decision from their host country as they have not been legally recognized as refugees.

²

The 1951 Convention defined refugees only in terms of ‘events occurring before 1 January 1951’—an expression that, for some contracting states, implied both temporal and spatial limitations. These limitations were removed by the 1967 Protocol Relating to the Status of Refugees; see UNHCR (2011). Further agreements have followed in order to extend the protection to cases beyond those covered by the Convention. These include the 1969 Convention of the Organization of African Unity, the 1984 Cartagena Declaration, and the European Union’s 2004 and 2011 Qualification Directives (Hatton, 2020).

³

Formally, Article 33(1) of the 1951 Convention provides that ‘No Contracting State shall expel or return (“refouler”) a refugee in any manner whatsoever to the frontiers of territories where his life or freedom would be threatened on account of his race, religion, nationality, membership of a particular social group or political opinion’ (UNHCR, 1994). In addition to non-refoulement, the 1951 Convention provides for ‘non-discrimination’ and ‘non-penalization’, as well as providing various rights regarding welfare, housing, education, and employment.

⁴

A protracted refugee situation refers to cases in which 25,000 or more refugees from the same nationality have been in exile for at least 5 consecutive years. By the end of 2020, UNCHR had an estimate of 15.7 million refugees living in protracted displacement (UNHCR, 2021).

⁵

Internally displaced people (IDPs) are different from refugees in that while refugees are under the protection of the international community, national governments are the ones responsible for IDPs. See the paper by Ibañez et al. (2022) in this issue for a thorough discussion.

⁶

Although not discussed in this issue, Colombia also hosts a large number of displaced Venezuelans, with close to two million by the end of 2021 (R4V, 2022).

⁷

For the US, there are also well documented examples of the role of refugees in increasing innovations and advances in R&D (Moser et al., 2014; Tam, 2020).

⁸

Previous work by the authors can be found in the project’s website: https://www.econforced.com/lamfor.

⁹

For broad reviews of the forced displacement literature in economics, see Verme and Schuettler (2021); Becker and Ferrara (2019); Dustmann et al. (2017); Ruiz and Vargas-Silva (2013).