Skip to content
The secondary migration of refugees resettled in the US

While the assumptions underlying the current refugee resettlement system in the United States (US) may have been true 35 years ago, the likelihood of an arriving refugee having no connections in the US diminishes every year. Every year the odds increase that an incoming refugee will have family or friends already living in the US, and advances in global connectivity have aided future refugees to keep in touch with former refugees. So incoming refugees, now more than ever, have strong connections and geographic preferences when arriving in the US. In recent years, homogeneous ethnic or cultural communities have begun to spring up all over the country, a phenomenon that refugee resettlement policy, which aims rather to ‘spread the burden’ of refugees across the country under a policy of dispersal in the initial placement of refugees, is actually designed to prevent.

It may seem unlikely that newly arrived refugees would spend their scarce financial resources on moving onwards when they have so many other challenging expenses. Refugees must repay a loan for the cost of their flight to their new home. Within six months they must begin paying rent. To do this they must quickly find a job. Picking up everything they own, again, and moving does not seem like something many refugees would choose to do. Yet, the data tell us, many do move.

In both fiscal year 2012 and 2013, the US accepted roughly 70,000 refugees; within a year of arrival, over 10,000 of the 2012 cohort and over 11,000 of the 2013 cohort had moved out of their initial resettlement community. However, these statistics only reflect what is reported to the US Office of Refugee Resettlement (ORR) and are probably underestimates of the true rates of secondary migration.

In some places secondary migration has massive consequences. For example, Minnesota accepted roughly 2,000 refugees in 2012 and 2013 but by the end of each fiscal year it had received over 2,000 more refugees through secondary migration, largely into the Minneapolis-St Paul metropolitan area.

Refugees move for the same reasons that any other person would move: to migrate towards increased job prospects, lower housing costs or better school systems; to live closer to friends or family members; or to take advantage of some geographical location or environmental amenity. It is rather the ways in which refugees differ from the average mover that make their secondary migration remarkable, as refugees generally have less wealth, initially do not have a full-time job, potentially do not speak English fluently, and have little knowledge about life in the US.

Finding out why refugees move on

When the US first began resettling refugees, the community in which they were initially assigned to live was as good as any other community in the country. Thirty-five years on, better alternatives exist – and refugees are increasingly aware of them. There are several factors in why refugees need to move on.

The incentive to remain silent: The most convenient time to gather information from refugees on their geographical preferences is during pre-resettlement interviews and meetings with them. There may, however, be an incentive for refugees to withhold information on their preferences due to a mistrust of bureaucratic officials or for fear of ruining their chances of actually being resettled in a third country. Remaining as amenable as possible may be a rational strategy for some refugees. Preferences may exist but the refugee may strategically withhold it believing that moving after arrival may well be a safer option compared with the risk of possibly losing the life-changing opportunity of resettlement. .

Gains versus losses: Refugees make choices based on gains and losses. Prior to arrival in the US, preferences tend to be general. Safety and resettlement anywhere is a gain, while remaining in a refugee camp or in danger due to violence is a loss. After arrival, however, the relative gains and losses from living in different locations shift and inform the decision to stay or move on.

Updated geographical information: Refugees may know about the whereabouts of their friends or family but may underestimate the distances in the US. Prior to their arrival, incoming refugees may wrongly assume that they will easily be able to visit relatives and friends already living in the US, simply because they will reside in the same country. Upon arrival refugees may desire to move in order to be nearer to their friends and family – and they may only realise this once they arrive in the country.

Network decision making: Relatively large groups of highly networked refugees now call various communities home. Some of these groups make migratory decisions as a group but arriving refugees may not have any idea where the group is planning on moving until after they arrive.

‘State-shopping’: Refugee resettlement programmes are extremely complex, with funds for assistance services coming from various budgetary streams, both public and private. Furthermore, eligibility for various services such as temporary assistance to needy families, Medicaid (the social health-care programme for families and individuals with limited resources) and employment assistance varies quite noticeably between states. This results in refugees engaging in so-called state-shopping in search of a location where they stand the best chance of becoming self-sufficient. In addition, stories and rumours circulate about where the best services and organisations are located. Friends who may have had a good experience with the services of a particular local resettlement organisation in a different community may influence an incoming refugee to migrate into this organisation’s service area in search of a similar experience.

Refugee policy

Any refugee resettlement practitioner will be quick to point out how important it is that refugees remain in their initial resettlement location for at least the first 90 days. Once a refugee moves, it is difficult for their services to be administered in the new community. For this reason most local voluntary agencies actively discourage refugees from moving soon after resettlement. Local agencies are acutely aware of the consequences of secondary migration on both the efficacy of the agency’s services and the welfare of refugees themselves. Refugees are promised special services for up to five years after arrival and a failure to adequately handle secondary migration places resettlement communities and refugees at risk. Some federal funding is provided for voluntary agencies that are dramatically affected by secondary migration; formulas for allocating such funds, however, are based on numbers of historical resettlement patterns rather than on projections of future patterns and often fall short of local needs.

Efforts to improve the dispersal of refugees resettled in the US have been made in recent years. In 2010 the ORR enumerated a number of principles to guide its services; in the explanation of these principles, however, secondary migration is mentioned only twice. Firstly, it is stated that “Appropriate placement and services from the onset is [sic] seen as a preventative measure against the challenges brought by secondary migration” and, secondly, there is a reference to the intention to develop a data system that can track secondary migration on from initial placements. Better placement techniques and additional data-informed decision making are certainly welcome improvements to the US refugee resettlement system but innovations are needed.

One such innovation often used to combat this issue is to present incoming refugees with a contract. In signing the contract refugees agree to report any information about the whereabouts of anybody they know who is currently living in the US. Additionally they agree not to move within the first 90 days of their arrival in the US under penalty of forfeiting their right to core services. This policy aims to change the incentives by rewarding refugees who share information but could penalise refugees who do not possess the information they might need in order to communicate what their geographic preferences will be once they arrive in the US.

The question that lies at the core of the challenges brought by secondary migration is whether the initial location should be a community where refugees are expected to settle or whether it is more of a receiving location, a launch pad, where refugees simply receive initial core services. The many implications of the answer to this question must be carefully considered. Given the current reality of resettling refugees and their secondary migration, perhaps the answer is to redesign resettlement to be more dynamic and to account for changing preferences.


Jeffrey Bloem
PhD Student, Department of Applied Economics, University of Minnesota

Scott Loveridge
Professor of Agricultural, Food, and Resource Economics, Michigan State University

This site is registered on as a development site. Switch to a production site key to remove this banner.