Outcomes for Internally Displaced Persons and Refugees in Low and Middle-Income Countries

The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.


Introduction
This chapter reviews the outcomes of those forcibly displaced due to conflict, violence and persecution and the factors that shape these outcomes.1It looks at refugees who flee across borders, as well as at those internally displaced in their own country.
The literature on outcomes of refugees and internally displaced persons (IDP) is nascent but growing.Much of the economics literature has focused on high-income countries because data are more readily available (Becker and Ferrara, 2019;Brell et al., 2020).Yet only onesixth of today's refugees and nearly no IDPs are hosted in those countries.A descriptive picture of refugee and IDP outcomes in low-and middle-income countries is emerging with the recent surge in representative cross-sectional data (Krishnan et al., 2020;Mejia-Mantilla et al., 2019;Pape et al., 2018).Efforts underway to increase panel data for those forcibly displaced in low-and middle-income countries will allow more rigorous, causal analysis in the future. 2n analyzing the outcomes of refugees and IDPs in their host communities, the literature has proceeded in two steps.First, it has sought to find the appropriate counterfactual to put the outcomes of refugees and IDPs in perspective, and in doing so has compared the outcomes of refugees and IDPs with other migrants and finally with residents of the host communities.Alternatively, studies have made comparisons with individuals who stayed behind.Where possible, such comparisons have been done over time, but only studies in high-income countries have been able to access panel or longitudinal registry data.Second, the literature has attempted to explain the observed differences or absence thereof, after controlling for observables.Two directions have been pursued.On the one hand, researchers have acknowledged that the conditions leading to the decision to flee create a selection process that in turn greatly influences the extent to which refugees and IDPs have been found to adapt to a new environment among their host communities.Where they settled is also not always random.On the other hand, while refugees, IDPs, migrants in general, and citizens all may live in the same location, they are often not subject to the same conditions and regulatory environment.Studies have thus focused on investigating how binding these constraints are for refugees and IDPs.The literature, therefore, has largely treated these two issues -migrant selection, and conditions and regulatory environments -as separable.
It is plausible, however, to assume that the treatment and regulation of refugee and IDP movements and market participation are in part determined by the size and composition of the refugee or IDP flow itself.This reality presents methodological challenges for establishing either or both the internal and external validity of past, ongoing, and future analyses of refugee and IDP welfare.
The literature highlights the important role of conflict dynamics in determining how those forcibly displaced are selected and how they compare to those who stay or other types of migrants.The costs of the journey, admission criteria, and networks at the destination also shape who goes where.Economic opportunities become important notably for secondary movements, beyond the first point of relocation.In addition to selection, those forcibly displaced are shaped by the shocks they experience before and during their displacement, including the loss of physical assets, human capital, and mental health.Their outcomes are further shaped by the conditions and policies they encounter at their destination.Demand for labor and potential mismatches between the labor they can supply and demand for that labor, social networks with co-nationals, and attitudes and perceptions of hosts all shape the economic and social integration of refugees and IDPs.Social and active labor market programs can help address the specific challenges refugees and IDPs face at their destination, and promote integration while limitations of their rights in terms of access to the labor market, the ability to move and settle freely, and predictability in terms of legal status have negative impacts.
Finally, a still under-researched question looks at longer-term outcomes when forcibly displaced persons either return to their countries of origin or are in durable or protracted situations abroad.The nascent literature needs to address another layer of selection -this time, into the return.First, results show that those who return usually tend to fare worse than those who stayed.Those remaining abroad and their children fare better over time if conditions are favorable.destinations (Ozden and Wagner, 2018), the bulk of refugees seek protection in a neighboring country.This thus makes the developing world the main host of forcibly displaced persons.

Forced displacement in numbers
As of 2020, 85 percent of the world's refugees were hosted in low and middle-income countries (UNHCR, 2021), although refugees are increasingly traveling longer distances to seek protection in a high-income country (Devictor et al., 2021).Low-income countries host 27 percent of all refugees, while middle-income countries host 58 percent (UNHCR, 2021).
The short-and long-term outcomes of refugees and IDPs will be driven by both characteristics of refugees and IDPs and the country or community hosting them and how these interact.These factors are discussed next.
3 Refugees and IDPs are no ordinary migrants Refugees and IDPs differ from those migrating primarily for economic reasons in at least two important dimensions: they leave their countries of origin in different circumstances and they might have experienced shocks and trauma that could affect their subsequent socioeconomic outcomes in their locations of destination.

Selection into migration
The decision to migrate is typically understood to be the result of the comparison of expected welfare in the origin and destination countries, net of (expected) migration costs (Roy, 1951;Sjaastad, 1962;Borjas, 1987).Unlike economic migrants, a forced migrant's first move is likely based on safety considerations and proximity instead of economic considerations.
Anecdotal evidence also suggests that forced migrants have little time to prepare for their move.For instance, half of the Syrian refugees surveyed in Kurdistan and more than onethird of those surveyed in Jordan had no more than one day to prepare for their move (Krishnan et al., 2020).
Whether, when, and where people move is influenced not only by awareness and experience of violence but also by the opportunities they have to move (Schon, 2019).Who leaves depends on conflict dynamics and varies between contexts in countries of origin (Ibáñez, 2014;Beber et al., 2021).If forced migrants are positively or negatively selected with regard to their human capital and welfare also differs between countries of destination.Compared to the Syrian labor force before the conflict, Syrian refugees in neighboring countries have lower levels of education and are less likely to have worked in a (high) skilled profession (Krishnan et al., 2020;World Bank Group, 2018).In contrast, the education level of Syrian refugees in Europe is high compared to their country of origin (Aksoy and Poutvaara, 2021;Brücker et al., 2020;Buber-Ennser et al., 2016).Furthermore, asylum seekers surveyed in Italy in 2016 are over-represented in the top quintiles of the income distribution in their countries of origin (World Bank Group, 2018).As for voluntary migrants, the costs of the journey, admission criteria, and networks at the destination are factors shaping these selection effects, in addition to opportunities to earn income at the destination versus the origin.

Shocks to physical and human capital and mental health
Refugees or IDPs migrating to seek protection most likely differ from other migrant populations because their decision to migrate is also associated with shocks to their physical and human capital and/or mental health.Such shocks either triggered their moves in the first place or affected them during transit.
Loss of physical capital Those forcibly displaced tend to have lower levels of assets compared to hosts and compared to their situation before displacement, cross-sectional data for IDPs and refugees from different low and middle-income countries show (Fiala, 2009;Krishnan et al., 2020;Ibáñez and Moya, 2010;Mejia-Mantilla et al., 2019;Pape et al., 2018;World Bank, 2019b).They usually cannot take their assets and seldom have enough time to sell them before their move; their assets are seized or they are not able to access assets left behind anymore.
Loss of human capital and skills Education and skills acquisition are likely to be disrupted through conflict and forced displacement.Children in Colombia recently internally displaced, for example, have significant gaps in educational accumulation and enrollment compared to other migrants and non-migrants.These gaps are larger than those of children who live in Colombian municipalities with high conflict (Wharton and Oyelere, 2011).Disruptions of work experience and longer periods of forced inactivity further deteriorate human capital.Refugees often face legal constraints to access the labor market (see later discussion).They also face more difficulties to access the labor market because they do not choose their destination primarily based on labor demand, usually have little time to prepare for the labor market at their destination, and often arrive as part of a larger group, that is, they are part of a large labor supply shock.At the level of the household, the separation from or loss of family members leads to higher dependency ratios.Further, refugee households are more likely to be headed by a woman than those of non-refugee households in various lowand middle-income destination countries (Mejia-Mantilla et al., 2019;Miguel et al., 2022;World Bank, 2019a;World Bank and UNHCR, 2020).
Deterioration in mental health Refugees and IDPs are also characterized by the traumatic events that either preceded their decision to migrate or along their journey away from their homes.Difficult conditions at arrival can reinforce negative mental health effects.Thus, studies usually find that refugees and IDPs suffer from worse mental health compared to the host population -although there is strong variation in the levels of incidence the studies find, due to the respective context and the methodology used (Porter and Haslam, 2005).
The experiences of violence and displacement also affect their aspirations, attitude toward risk, and willingness to make investments specific to the host country (Schuettler and Caron, 2020;Chiovelli et al., 2021).

Characteristics of host communities
Beyond the attributes of migrants discussed earlier, critical determinants of the welfare of refugees and IDPs include the characteristics of the communities that host them.These can be exogenous or themselves the outcome of the incoming flow of forcibly displaced individuals.
In host communities, an inflow of refugees and IDPs can affect labor market outcomes, prices, provision of social services, availability of infrastructure and natural resources, crime rates, and social cohesion, among others (Becker and Ferrara, 2019;Verme and Schuettler, 2021).
The perceptions of such impacts do not always align with the actual (aggregate) impact (Kreibaum, 2016).Nonetheless, perceptions influence attitudes and policies toward those forcibly displaced.In investigating the conditions under which the forcibly displaced are able to successfully integrate into their host communities, the discussion first examines the socio-economic environment where refugees and IDPs end up, in terms of the labor market, social networks and attitudes and perceptions of hosts.It then looks at the roles of policies and rights in encouraging or impeding social integration.

Socioeconomic conditions 4.1.1 Labor market conditions
Refugees who arrive at the country of destination during a recession or are placed in areas with lower employment rates within the country have worse labor market outcomes, which then persist over time (Aslund and Rooth, 2007;Azlor et al., 2020;Barth et al., 2004;Braun and Dwenger, 2020;Fasani et al., 2022;Godøy, 2017).The literature shows this correlation consistently but is focused on high-income countries due to data availability.The results are corroborated by a larger literature on migrants more generally.Potential channels for the persistence of these worse labor market outcomes are scarring effects at the individual level.In addition, for those placed in areas with worse employment conditions within countries, the reason might be persistent differences in employment conditions combined with limited geographical mobility of those forcibly displaced, even where they are allowed to settle elsewhere.
Economic outcomes are also worse when the skills of those displaced do not match those demanded in the labor market.For example, a study of refugees in Switzerland found that the greater the language distance between the country of origin and the area asylum seekers were assigned to, the lower the positive impact of an open labor market access policy on the employment rate of asylum seekers (Slotwinski et al., 2019).These results were corroborated by another study on Switzerland that showed that, controlling for other observable characteristics, refugees randomly placed in areas with languages most similar to their own were more likely to find employment (Auer, 2018).A simulation for the United States and Switzerland using machine learning to place refugees finds that it would improve the employment outcomes of refugees by 40 percent to 70 percent through better matching compared to the current procedure of randomly assigning where to settle them without taking potential labor market outcomes into account (Bansak et al., 2018).

Networks with co-nationals
Beyond labor demand for their skills, social networks at the destination are important for economic and social integration.A sizeable literature looks at the role of networks with co-nationals with a focus on high-income countries.It has been widely documented that the forcibly displaced seek a settlement in communities where there already is a sizeable population that shares common social attributes such as religion, ethnicity, or language.
This finding is quite general and has been documented in both developed countries and developing countries.Based on the work of Altonji and Card (1991); Card (2001), many authors have used that feature to predict where refugees and IDPs would settle either across countries or within countries (Altındag et al., 2020;Caruso et al., 2019;Morales, 2018).
The strategic location decision has been shown to yield some positive results in terms of integration into the local labor market, highlighting the importance of social networks in determining refugees' labor market outcomes.Only a few studies focus on networks with hosts, all of which find an association with positive labor market outcomes (Cheung and Phillimore, 2014;De Vroome and Tubergen, 2010).The remaining studies all focus on networks with co-nationals, usually defined as those residing in the same area.Exploiting random dispersal policies, several studies for high-income countries find positive impacts of co-nationals on earnings or employment (Damm, 2009;Damm and Rosholm, 2010;Martén et al., 2019;Edin et al., 2003).
However, to the extent that labor markets are thin, a large inflow of refugees or IDPs over a short period of time might end up crowding out labor markets, adversely affecting those arriving at the same time, as well as natives and previous immigrants, and hence dampening the positive social network externality (Beaman, 2011).More generally, the quality of the network matters for labor market outcomes (Battisti et al., 2022;Dagnelie et al., 2019;Parsons et al., 2020;Stips and Kis-Katos, 2020).

Attitudes and perceptions of hosts
Besides labor market conditions and networks with co-nationals, the economic and social integration of refugees and IDPs can be hindered or facilitated by the host community perceptions of and attitudes towards the forcibly displaced (Aksoy et al., 2020;Bailey et al., 2022).These attitudes and perceptions affect the ability of refugees and IDPs to build up social networks with hosts that are important for integration.Discriminatory hiring practices (Carlsson, 2010;Loiacono and Silva Vargas, 2019) and teacher biases in grading (Alesina et al., 2018b) are examples of other key channels through which perceptions and attitudes can influence outcomes.A broader literature shows how discrimination and stereotypes can lead to expectancy effects and self-fulfilling prophecies (Bertrand and Duflo, 2016).Perceptions and attitudes also indirectly impact refugee and IDP outcomes by influencing policies.
Research from different social sciences has unveiled some of the determinants of host community perceptions of and attitudes towards refugees and IDPs. 4 Studies looked at whether these were shaped by economic, political, and cultural conditions.In general, attitudes are shaped by perceptions of the economic impact (including labor market competition, congestion of public services, and fiscal impacts), and concerns about the cultural and security impact of immigrants on the in-group (i.e. the host nation or respective ethnic group).
Economic self-interests seem to play a less important role (Hainmueller and Hopkins, 2014;Valentino et al., 2019).Economic downturns seem to be correlated with more negative attitudes toward immigrants.Further, the speed and spatial concentration of the inflow seem to matter more than absolute numbers (Albarosa and Elsner, 2022).Those directly exposed to transit migration of refugees have more negative attitudes toward refugees, likely because they have less opportunity for repeated interaction and feel a loss of control (Ajzenman et al., 2020;Steinmayr, 2021;Hangartner et al., 2019).
Perceptions and attitudes also depend on the characteristics of refugees and IDPs.Those in need of support and protection are usually perceived more positively than other migrants (Adida et al., 2019;Alrababa'h et al., 2021;Bansak et al., 2016;von Hermanni and Neumann, 2019).Besides favoring those who are most vulnerable (often women and children), the same studies also consistently show that hosts seem to have a preference for refugees from the same culture or religion and those with higher employability.Inflows of those considered as more culturally distant raise more cultural concerns.But being too close can also trigger resistance (Adida, 2011).Hosts tend to misperceive the number and composition of immigrants more generally, which is associated with more negative attitudes (Alesina et al., 2018a).
In addition, individual characteristics of the host population shape their perceptions and attitudes.While education is usually negatively correlated with anti-immigration senti-ments more generally in high-income countries (Hainmueller and Hopkins, 2014), this does not necessarily seem to be the case in low-income countries (Mayda, 2006) and for antirefugee sentiments (Adida et al., 2019;Lergetporer et al., 2021).There is little evidence of attitudes towards immigrants being strongly correlated with personal economic circumstances (Alrababa'h et al., 2021;Hainmueller and Hopkins, 2014;Valentino et al., 2019).
There is a tendency for more openness among younger generations and among those who are more politically liberal (Dempster et al., 2020).Openness to interact with and apply values to others outside the own in-group differs between host countries, but also between individuals living in rural and urban areas (Dustmann et al., 2019).Prejudices, ethnocentrism, and adherence to stereotypes are clearly connected to stronger anti-immigration sentiments (Hainmueller and Hopkins, 2014) and to preferences for culturally more similar immigrants.
The impact of previous experience with conflict and violence is ambiguous: it might increase empathy with those forcibly displaced, but could also lead to a higher perceived threat from out-groups.Existing evidence is mixed (Ghosn et al., 2019;Hartman et al., 2021;Hong et al., 2020).Previous experience with displacement or migration seems to have a more clearly positive impact (Adida et al., 2019;Barron et al., 2021;Hong et al., 2020).
While these perceptions and attitudes of the hosts are potentially driven by exogenous attributes of the migrant population as well as exogenous factors in host communities, they might also be influenced by the endogenous outcome: that is, how well migrants end up integrating.Causality can also go both ways for the host communities' politicians, media, and civil society, which shape attitudes and perceptions with their narratives, but also react to existing attitudes and perceptions.
Finally, an increasing body of literature suggests that perception and attitudes are actually malleable, and do not only change slowly over time, from one generation to the next, or through shocks.These studies look at how attitudes and perceptions change through contact, perspective taking, personal narratives, and stressing commonalities as well as through information.A variety of these studies are influenced by the contact theory developed by Allport (1954).In their meta-analysis of experimental studies to reduce prejudices against different groups, including refugees and migrants, Paluck et al. (2021) find the effects on reducing prejudices against refugees and other immigrants to be strongest from extended or imagined contact, followed by peer influence.Their review, however, cautions against expecting large impacts.They find that overall, more precisely measured studies find much smaller effects, and point out that studies have so far focused on short-term impacts.Notably, in lowand middle-income countries, there is still only very limited experimental evidence about whether intergroup contact actually improves locals' relations with displaced persons (Zhou and Lyall, 2022).Studies using observational data, such as Ghosn et al. (2019), must be interpreted with caution, given that a positive correlation between contact and lower prejudices might be due to less-prejudiced individuals seeking contact.Future research will also need to explore the importance of the conditions defined by Allport (1954) under which the interactions between in-group and out-group individuals would increase empathy and integration: that is, that (i) both groups must share status, (ii) and the same goals or interests, (iii) both groups need to live in a cooperative environment rather than a competitive one and (iv) they need to benefit from institutional support: that is, operate under a well-defined set of norms, laws, and regulation (Paluck et al., 2019).Studies so far only show that proximity alone (Zhou, 2018) and brief exposure in the framework of transit migration do not seem to amount to the same as contact (Ajzenman et al., 2020;Steinmayr, 2021;Hangartner et al., 2019).
Related to studies inspired by Allport's contact theory are studies that look at the impacts of perspective-taking and providing personal narratives, as well as studies stressing commonalities (in terms of conflict and displacement and other dimensions).Studies focusing on forcibly displaced populations find positive but not always significant or large impacts (Adida et al., 2018;Alan et al., 2021;Audette et al., 2020;Dinas et al., 2021;Hong et al., 2020;Lazarev and Sharma, 2017;Rodríguez Chatruc and Rozo, 2021).While the provision of facts to correct misperceptions has shown to change attitudes in certain settings, notably if combined with information about characteristics and impacts of immigrants and refugees (Grigorieff et al., 2020;Haaland and Roth, 2020;Lergetporer et al., 2021), other studies find no or mixed results (Adida et al., 2018;Alesina et al., 2018a;Barrera et al., 2020).Once attitudes are formed it seems they are not easily changed and contact and narratives are more effective than facts in this regard.Public investments made for refugees which benefit hosts can influence their attitudes in a positive way (Quattrochi et al., 2019;Lehmann and Masterson, 2020;Baseler et al., 2021;Zhou et al., 2022;Valli et al., 2018), but this is not always the case (Kreibaum, 2016).

4.2
The role of policies 4.2.1 Social programs and active labor market policies As they are seeking protection from violence and persecution, refugees and IDPs do not necessarily choose their destination to maximize their labor market contributions.The decision to migrate is also often associated with loss of income, assets, and human capital.Social programs and active labor market policies are thus lifelines for these vulnerable populations.
A growing literature has looked at the impacts of such programs and policies directed at refugees and IDPs.These include social transfers (cash or in-kind, whether one-off or repeated), healthcare and psychosocial support, skills training, and other active labor market policies.
Social transfers can compensate at least in part for the loss of assets and income or even lead to an increase in assets for those households that had very few assets before their displacement (Fiala, 2015;Quattrochi et al., 2019) -at least while the program lasts (Altındag and O'Connell, 2021).They allow households to cover their basic needs, invest in education and health, reduce child labor, and lift financial constraints that prevent them from searching for higher-quality jobs and taking more risks; they can foster social integration if administered correctly (Aker, 2013;Aygün et al., 2021;Caria et al., 2020;Chaaban et al., 2020;Hidrobo et al., 2014;Lehmann and Masterson, 2014;MacPherson and Sterck, 2021).
Labor-intensive public works seem to have similar, short-term impacts for refugees and IDPs but there is very little evidence (Lombardini and Mager, 2019).In contexts where refugees and IDPs are forced to remain idle, employment can significantly improve psychosocial wellbeing beyond the impacts of cash alone (Hussam et al., 2021).
Therapy and other socio-emotional support interventions have a very good track record in improving the mental well-being of refugees and IDPs (Acarturk et al., 2016;Bolton et al., 2016;Islam et al., 2022;Knaevelsrud et al., 2015;Neuner et al., 2004Neuner et al., , 2008)).Different types of skills training (including language training) are more successful in improving employment rates and wages of refugees and IDPs in the long term and when combined with other interventions (Schuettler and Caron, 2020).The training needs to be adapted to labor demand in the location where those forcibly displaced reside.It also must take into account the aspirations of the target group, as well as the legal framework that affects their access to the labor market.In terms of other active labor market policies, job matching and counseling interventions have shown positive results for refugees in high-income countries (Aslund and Johansson, 2011;Battisti et al., 2019;Joona and Nekby, 2012).The limited evidence for low-and middle-income countries is less clear (Caria et al., 2020).The same is true for wage subsidies (Clausen et al., 2009;Heinesen et al., 2013).

Rights granted to refugees and IDPs
In an attempt to control the flow and composition of refugees seeking protection on their own territories and manage the actual or perceived negative externalities stemming from an influx of migrants, governments have often chosen to curtail the rights of refugees.In some countries, the rights of IDPs have been limited as well.While de jure refugee and asylum policies have become more restrictive in high-income countries over time, they have become more liberal in low and middle-income countries since the 1950s (Blair et al., 2021).5 .
The welfare cost of these measures can be assessed when they are rescinded.Providing Venezuelans with a legal resident status increased their consumption and income and decreased their food insecurity compared to those residing irregularly in Colombia (Ibánez et al., 2022).Evidence from high-income countries shows positive impacts from limiting or shortening employment bans for asylum seekers (Marbach et al., 2018;Slotwinski et al., 2019) and reducing wait time for recognition of refugee status (Hainmueller et al., 2016;Hvidtfeldt et al., 2018), as well as a positive correlation between acquiring citizenship and better economic outcomes (Bakker et al., 2014;Bevelander and Pendakur, 2006;Govind, 2021).At the same time, results for policies to disperse populations of refugees geographically within high-income countries consistently show that not allowing refugees freedom of movement and settlement has negative impacts on their economic outcomes (Aksoy et al., 2020;Azlor et al., 2020;Fasani et al., 2022;Martén et al., 2019;Edin et al., 2003).Similarly, living in camps is associated with higher inactivity in Türkiye and lower earnings in Jordan and Iraq (which are offset by subsidies offered in camps), as well as the quality of life in Jordan beyond income (Ginn, 2020;Kayaoglu and Erdogan, 2019;Obi, 2021).

Longer-term outcomes
In the longer term, refugees either return to their origin countries or settle in their host communities (integrating locally, resettling to a third country, or staying in so-called protracted situations). 6The median duration of exile was five years as of 2018, Devictor and Do (2017) find.The number of returns of IDPs is higher than for refugees, but the count for IDPs is less reliable.The outcomes of those who return are discussed first, followed by the outcomes of those forcibly displaced and their children who stay in host communities.

Returning to the origin community
Studies on factors influencing return decisions show the importance of the security situation, economic opportunities, property rights, networks, and services in the origin community, and the quality of information about these factors (Arias et al., 2014;Camarena and Hagerdal, 2020;Stefanovic et al., 2015;Weber and Hartman, 2021).Recent investigations find that conditions in the origin country are the main factor for intentions to return and actual returns of Syrian refugees in neighboring countries as well as Germany (Alrababa'h et al., 2020;Beaman et al., 2022;Kaya and Orchard, 2020).While some previous studies find that a better situation at the destination is associated with lower return intentions (Arias et al., 2014;Balcilar and Nugent, 2019;Stefanovic et al., 2015), more recent studies show that the situation at the destination does not play a role or is even positively correlated with return (Alrababa'h et al., 2020;Beaman et al., 2022).When the forcibly displaced are better off at their destination they have more to lose when returning but at the same time, they have the resources to actually return.This explanation is in line with findings showing that those with higher education are more likely to return (Kaya and Orchard, 2020;Stefanovic et al., 2015).
A few studies look at the outcomes of forced migrants after their return to their origin communities.In addition to the selection aspects discussed earlier, they need to address the selection into return.While a study on Rwanda finds that returnees have higher returns to on-farm labor, likely due to higher motivation (Kondylis, 2008), studies for Bosnia (Kondylis, 2010), Burundi (Fransen et al., 2017;Verwimp and Muñoz-Mora, 2018), Sudan (Abdel-Rahim et al., 2018), and Uganda (Fiala, 2015) all show that previously displaced returnees are worse off compared to those who stayed.Reasons might include an inability to recover previous assets, deterioration of human and social capital specific to the origin community, and socio-emotional challenges to reintegration.Some studies, however, show that the displaced might have gained human capital and adopted beneficial new habits while abroad (Abdel-Rahim et al., 2018;Fransen et al., 2018).

Staying in host communities
As the number of returns are low, many refugees and IDPs end up staying in host communities.Their outcomes are shaped by the selection effects, the impacts of the forced displacement experience, the socio-economic environment as well as the policies discussed earlier.The discussion focuses on economic outcomes followed by the human capital outcomes of the next generation.

Economic outcomes
Refugees and IDPs often fare worse when compared with hosts and other migrants at destination.Refugees and IDPs are poorer than hosts in many destination countries.IDPs in Iraq, North East Nigeria, South Sudan, Sudan, and Somalia as well as refugees in Ethiopia all face a higher incidence of multi-dimensional poverty than hosts, with a difference ranging from 15 percentage points to more than 30 percentage points (Admasu et al., 2021;Noumedem Temgoua et al., 2020).In Jordan, only 12 percent of Syrian refugees were neither poor at the time of measurement nor vulnerable to poverty in the near future (Verme et al., 2016).
Evidence from IDPs in Iraq suggests a causal relationship between forced displacement and poverty (Noumedem Temgoua et al., 2020).Forced displacement has also been negatively associated with consumption levels (Fiala, 2015;Ibáñez and Vélez, 2007), wages and employment (Schuettler and Caron, 2020), and food security (Pape and Sharma, 2019).Where conditions are favorable, those forcibly displaced can close at least part of this gap over time, as evidence on labor market outcomes from high-income countries shows (Schuettler and Caron, 2020).

Health and education for the next generation
Various studies investigate long-term oucomes in terms of health and education for the second generation, which comprises either those who arrived at a young age as refugees or IDPs or were born to forcibly displaced parents.When examining these outcomes, studies need to take into account not only selection effects and non-random settlement, but also non-random onward and return migration.Refugee and IDP children benefit if they find themselves in a destination with access to better healthcare and other health enhancing factors.A study on the airlift of Ethiopian Jews to Israel shows that children who benefitted from earlier access to better prenatal health care in Israel had better educational and labor market outcomes later in life (Lavy et al., 2021).Poor mental health of primary caregivers due to displacement can have negative impacts on their children.A study sampling Syrian refugees in Jordan, for instance, found that more than half were likely to be depressed, and the mental health of refugee mothers was highly correlated with that of their children (Miguel et al., 2022).A randomized control trial among Rohingya mothers and their children in Bangladesh refugee camps showed that a multifaceted psycho-social intervention not only had positive impacts on the mental health of mothers but also on their children and improved child development outcomes more generally (Hussam et al., 2021).7 An examination of educational outcomes using descriptive statistics (Mejia-Mantilla et al., 2019;Pape et al., 2018;Pape and Sharma, 2019) and studies controlling for observables (Wharton and Oyelere, 2011) shows that refugee and IDP children in low and middle-income countries tend to have worse educational outcomes than hosts.But this is not always the case, notably where the international community provides services in poor and remote areas (Nguyen et al., 2021).Who flees where, the displacement experience itself, and the situation at the destination all influence outcomes.Further analysis would need to identify which role selection effects play as opposed to disruptions due to displacement, difficulties in adapting to the new school system (including in terms of language), or differences in financial barriers and the proximity and quality of schools refugee and IDP students can attend compared to host students and those who stayed behind.In addition to socio-economic differences between those forcibly displaced and their hosts, age at arrival also matters (Kirdar et al., 2021), as corroborated by evidence for migrant children more broadly.Due to the displacement experience, forcibly displaced parents might prefer investments in education instead of assets, because human capital is mobile (as identified by the so-called uprootedness hypothesis).Evidence from Mozambique and Poland shows that displaced children benefit from higher investments in education compared to those who stayed (Becker et al., 2020;Chiovelli et al., 2021).Such changes in preferences can, however, materialize only if conditions at destinations are favorable.Along these lines, descriptive statistics from refugees in Ethiopia show that primary school enrollment is higher but secondary school enrollment is lower compared to the country of origin (Pape et al., 2018).In the short term, displaced parents in Bosnia-Herzegovina invest less in their childrens' education compared to stayers (Eder, 2014).

Conclusion
Research on forced displacement has dramatically expanded in the past ten or so years.The literature has progressively moved beyond issues specific to forcibly displaced populations (such as the impact of humanitarian and other social assistance programs for refugees and IDPs) and has also focused on issues common to the larger field of migration research (such as socioeconomic integration, host community perceptions, and long-term outcomes).Limitations on the freedom of refugees to settle and move freely have been exploited to address endogeneity concerns linked to the non-random settlement of migrants so studies can more rigorously answer questions of interest to the broader migration literature.While the availability of data is still restricting the geographical scope of analyses, the literature has made significant progress in understanding not only how well refugees and IDPs fare at their destinations, but also the conditions under which their outcomes are improved.Future research will benefit from the increasing data collection efforts in low-and middle-income countries, notably if more panel data are collected, as planned.A greater focus on IDPs is warranted in the literature to counterbalance the current focus on refugees.
About 90 percent of persons forcibly displaced due to conflict, violence, and persecution are in the developing world because most of them are either internally displaced or, when they cross a border, they have gone to a nearby country.IDPs are notoriously hard to count.Definitions vary and it is unclear when someone ends up being internally displaced.In 2021, 53.2 million were internally displaced due to conflict and violence in 59 countries and territories, of which 99 percent were low and middle-income countries, the Internal Displacement Monitoring Centre (IDMC) estimates.(Internal Displacement Monitoring Centre, 2022). 3The number of IDPs has nearly doubled over the last decade (figure 1).Source: Internal Displacement Monitoring Center (IDMC) -Data accessed May 4, 2022 Note: IDPs = internally displaced persons.

Figure 1 :Figure 2 :
Figure 1: The number of internally displaced persons has nearly doubled since 2009