Taking Politics Out of Refugee Policy: Improving US Resettlement for Those Most in Need

Smith, Josh T.; Idso, Rachel; Schulz, Christine

The United States refugee program began in 1980 and has resettled about 3.7 million people since then.¹ That is approximately 85,000 refugees each year. Yet in recent years the US has admitted fewer refugees. In 2021, only 15,000 refugees were resettled, and in 2022, the US resettled fewer than 26,000.² Why has refugee resettlement fallen so much in recent years?

The global number of refugees has not similarly fallen. Instead, it has grown: there are about 27 million refugees today, according to the United Nations.³ The reduction in resettlement to the United States has taken place despite research that makes it clear that there are large economic and security benefits to admitting refugees.⁴

This report examines why the US has admitted fewer refugees in recent years and how the US compares to similar countries in terms of refugee resettlement. It also considers options for admitting greater numbers of refugees. Our central finding is that the US can both match and exceed other countries in refugee resettlement. Instead of setting political goals, like the current goal of 125,000 (which the country is unlikely to reach), the United States should be changing the way refugees are admitted, which would allow us to set our sights higher—to at least 250,000. This would render the US resettlement rate similar to the rates of countries such as Canada and Germany.

Our major findings and policy proposals are as follows:

The US resettles relatively few refugees compared to similar countries after we adjust for the much larger US population.
Relying on a needs-based refugee resettlement approach would imply a much greater level of resettlement by the US.
Restoring and expanding the US private refugee resettlement capacity could bring in many more refugees at little public cost.
Adopting state-based resettlement options may avoid political debates about refugee levels by letting states opt in to or out of additional resettlement.
The US is already piloting a refugee matching system that could modernize and simplify the resettlement process so that the country can take in more refugees.

This report begins with an explanation of how the US refugee system currently works and then compares the US to other countries. It concludes with recommendations for increasing resettlement levels in the United States. Raising US resettlement levels to be comparable with those of other countries will be a long-term project. It’s one that will generate large and important benefits for the country and for the refugees themselves if it is successful.

How does the US refugee system work?

Refugee resettlement is a global challenge. There are currently 27 million refugees in the world.⁵ Recent conflicts in the Middle East have dramatically increased the number of refugees worldwide. The Syrian civil war that began in 2011 has especially contributed to the current refugee crisis.⁶

The United States cooperates with international groups to evaluate refugees for resettlement and to coordinate aid efforts. Since 2016, the United States has taken in approximately 190,000 refugees—about 27,000 a year—though it often sets much higher goals. President Biden set the refugee ceiling for fiscal years 2022 at 125,000 refugees.⁷ However, the US only resettled 25,465 refugees, leaving nearly 100,000 slots unfilled.⁸ The 2023 goal is again 125,000.⁹

This section provides an overview of the refugee resettlement system in the United States and explains how it works within the international system. Specifically, the section explains who counts as a refugee, how many refugees the United States admits in relation to the worldwide refugee population, what the process is for a refugee to come to the United States, and what combination of public and private support is available to refugees once they are in the country.

Who is a refugee?

Under both US and international law, a refugee is someone with a well-founded fear of persecution on the basis of a protected characteristic. Protected characteristics include religion, race, and nationality.¹⁰ Additionally, belonging to a specific social group or holding certain political opinions can qualify someone for refugee status. So family members persecuted for their religion can qualify as refugees, as can pro-democracy advocates who face discrimination because of their political activities in countries with authoritarian regimes.

Individuals who flee their country because of a fear of persecution qualify as refugees only once they have gone through the US refugee status determination process. This designation process is a key difference between refugees and asylum seekers. Asylum seekers go from their home country directly to a permanent host country. Refugees, on the other hand, go from their home country to a temporary country of asylum, where they go through a formal designation process, and then they go on to a permanent host country.

The United States grants both refugees and asylum seekers protection. This means that individuals who are fleeing persecution and are at the border or already in the United States are granted asylum status and don’t need to seek refugee status.¹¹ Refugees who come to the United States, by contrast, previously sought asylum in a foreign country or were displaced and designated as refugees.¹² Not every asylum seeker will meet the criterion to be a refugee.

The legal framework for refugees under US law is documented in the Immigration and Nationality Act of 1952. On an international level, the Convention and Protocol Relating to the Status of Refugees constitutes the legal framework for international refugee law. By signing the 1967 refugee protocols, the United States and 148 other countries committed to upholding the human rights of refugees.¹³ The Convention Relating to the Status of Refugees defines a refugee as anyone who “owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion is unable or, owing to such fear, is unwilling to avail himself of the protection of [his home] country.”¹⁴

How many refugees exist in the world?

According to data from the United Nations High Commissioner for Refugees (UNHCR), as of 2019 there were 89.3 million forcibly displaced people worldwide, of which 27.1 million are refugees and 4.6 million are asylum seekers.¹⁵ In addition to refugees and asylum seekers, UNHCR also recognizes internally displaced persons and Venezuelans who are displaced abroad and cannot return to their home country but haven’t formally sought asylum. These groups, as seen in figure 1, constitute the displaced people in the world.

While the number of refugees worldwide has increased dramatically in the past decade, the United States has seen a decline in the number of refugees admitted to the country compared to pre-2015 levels. Figure 2 presents the number of refugees admitted to the United States from 2001 to 2022. It shows the steep decline in US refugee resettlement. Throughout the paper, we use UNHCR data on refugee resettlement to achieve consistency between US and international data in our comparisons. But in figure 2 we also include the US counts of refugees resettled to display the small differences between the US data and the UNHCR data. The two data sets (one from UNHCR and one from the Worldwide Refugee Admissions Processing System, or WRAPS) differ slightly, but not in substantial ways.

The minor differences shown in figure 2 arise for several reasons. The United States posts data for the fiscal year, whereas UNHCR uses data for the calendar year. Furthermore, there might be refugees who come to the United States through means other than a UNHCR referral—for example, by direct referral at US embassies.¹⁶ These refugees would not be included in the UNHCR numbers.¹⁷

How do refugees get to the United States?

The refugee admission and resettlement process in the United States broadly follows this five-step outline:

The US president sets the number of refugees to be accepted.
Potential candidates are referred to the US Refugee Admissions Program by a US embassy, by UNHCR, or by a designated nongovernmental organization.
The US Department of State’s Resettlement Support Centers determine the eligibility of referred individuals for refugee status.
Refugees are brought to the United States.
Refugees receive aid and assistance from public and private organizations in the United States and then can apply for legal permanent residence and, later, for citizenship.

The decline in the number of refugees resettled in the United States since 2000 occurred because each year the executive branch has chosen to admit fewer refugees. The executive branch in the United States has the authority to set annual refugee admission numbers after consulting with Congress. The consultation considers the current humanitarian needs of refugees and the US capacity to admit refugees.¹⁸ The president’s administration drafts and the president signs the Presidential Determination on Refugee Admissions, which sets the refugee cap for the upcoming fiscal year. The refugee caps from 2000 to 2022 are shown in figure 3.

Presidents have set the refugee cap at lower levels for many reasons. One is the effect of 9/11 on immigration policy and the political environment.¹⁹ After the 2001 attacks, President George W. Bush halted admissions for two months.²⁰ The FBI also added new security-check requirements to the process, which slowed refugee resettlement.²¹ In the two years following 2001, less than half of the refugees allowed by the refugee cap were admitted to the country.²² Though President Barack Obama returned the program to its previous levels, President Donald Trump again lowered the refugee cap.

In 2020, the Trump administration set the refugee cap for fiscal year 2021 at 18,000.²³ The following year, the Biden administration originally set the refugee cap even lower, at 15,000. This likely reflected administration officials’ beliefs about the capabilities of US resettlement infrastructure. But later the administration released an Emergency Presidential Determination on Refugee Admissions and an accompanying Memorandum for the Secretary of State, which increased the refugee cap for 2021 from 15,000 to 62,500.²⁴ As figure 3 shows, raising the refugee cap to 62,500 in 2021 brought it back to a more comparable level with the limits between 2000 and 2015.

Notably, the United States rarely reaches the ceiling stipulated by the president. Only six times between 2000 and 2022 did the United States resettle as many refugees as allowed under the cap.²⁵ But the resettlement number is often close to the maximum allowed, as shown in figure 3.

Recent years, however, have seen dramatic shortfalls in resettlement numbers. In fiscal year 2021, when the government was allowed to resettle up to 62,500 refugees, only 11,411 were resettled. That is, the United States resettled only around one out of five of the refugees that it could have.²⁶

The gap between the resettlement ceiling and actual resettlement numbers happened because the State Department downsized the refugee resettlement infrastructure in 2017. From that year until 2021, about a third of resettlement offices closed.²⁷ In 2021, the US refugee program had about 62 percent of the offices it had in 2016. When the Biden administration began, resettlement agencies did not have the resources to resettle many more than 11,000 refugees.²⁸

Resettlement in 2022 followed a similar trajectory. Only 25,465 refugees were resettled according to the US data and about half of that according to the UNHCR’s numbers. That is again only about one in five of the refugees who could have been resettled under the 125,000 goal. Though the figures exclude Afghans and Ukrainians brought to the US, because they were brought outside formal refugee channels.

Once the level of refugee admissions has been determined by the president, the United States works with UNHCR to select refugees.²⁹ This starts with UNHCR designating individuals as refugees on the basis of their claims under international law.³⁰ UNHCR, US embassies, and nongovernmental organizations identify individuals who are in need of resettlement and begin selecting some for resettlement in the United States.

These individuals are referred to the US Refugee Admissions Program and categorized into one of three priorities:

Priority 1 (P-1) includes individual referrals that meet the definition of a refugee, meaning that resettlement is a desirable long-term solution. The majority of P-1 cases are referred to the US by UNHCR.
Priority 2 (P-2) consists of specific predetermined groups. These groups include religious minority groups in Eurasia, the Baltics, and Iran, as well as Iraqis associated with the US.
Priority 3 (P-3) is a pathway for family reunification. Individuals of specified nationalities whose immediate family members came to the US as refugees can be given P-3 status, allowing them to come to the US.³¹

Before these referrals receive the rights of refugees, their refugee status needs to be determined to ensure that their claims qualify them as refugees under international law. Refugee status determination is primarily conducted by the United States Citizenship and Immigration Services.³² The organization making the determination of status collects personal data, immigration data (previous visa refusals and immigration law violations), and biometrics. This information is used to conduct background checks and assess the applicant’s criminal history. The entire process can take 18 to 24 months to complete.³³

Resettlement submissions are given varying levels of priority depending on the urgency of each individual case.³⁴ This recognizes that some refugees may need to be granted refugee status and referred to a host country for resettlement sooner than others.

A refugee screened and then selected to come to the United States receives a brief orientation about the culture of the United States before being brought to the country.³⁵ Typically, the International Organization for Migration coordinates flight arrangements.³⁶ It receives the funds for travel expenses from the State Department’s Bureau of Population, Refugees, and Migration (PRM).³⁷ The International Organization for Migration uses these funds to issue refugees an interest-free loan to pay for their travel expenses.³⁸ Since this is a loan, refugees repay the funds. Other than the implicit subsidization of their travel through the interest-free loan, refugees pay for their own travel to the US. The US government is reimbursed for refugees’ travel expenses.

How are refugees resettled?

Once refugees are approved for resettlement, they are sponsored by a nonprofit resettlement agency. This agency determines their placement in the United States and provides initial resettlement services after their arrival.

There are two major phases of federal support for refugees’ resettlement after they come to the United States.³⁹ In the first phase, the PRM provides resettlement agencies with a one-time payment per refugee to assist with initial support services. The second phase consists of programs administered by nongovernmental organizations and states and delivered in partnership with (and funded by) the Department of Health and Human Services Office of Refugee Resettlement.

Nine major resettlement agencies determine placement and determine which agency will sponsor incoming refugees. These voluntary agencies meet weekly to review biographical information and other case records sent by the Department of State’s overseas Resettlement Support Centers.⁴⁰ The VOLAGs are required to consult with—and take recommendations from—local authorities, state governments, and local nonprofits before resettlement.⁴¹ Placement is largely dependent on whether refugees already have close family and friends in the United States, on opportunities for employment, and on the presence of strong community services.⁴²

After placement, the sponsoring resettlement agencies’ local affiliates provide initial services. These services include housing assistance, enrollment in employment services, school registration, applications for social security, connecting refugees with language or social services, and other basic services.⁴³ The PRM’s one-time payment to resettlement agencies covers a majority of the costs for these services. The PRM expects sponsoring agencies to supplement federal funding with significant cash or in-kind resources for the initial three-month period of resettlement.⁴⁴

After this 30–90 day period, the Office of Refugee Resettlement takes over from the PRM, working with states and local affiliate resettlement agencies to provide longer-term case management and medical assistance in addition to language services, employment services, and social services.⁴⁵ The Office of Refugee Resettlement has several programs that states and nongovernmental organizations administer to provide refugees with support. These include the Cash and Medical Assistance program, which provides funds for medical costs; the Ethnic Community Self-Help program; and specialized programs such as the Individual Development Accounts program, which helps refugees save for a car, a home purchase, starting a business, or postsecondary education.⁴⁶

Refugees who have been relocated to the United States are required to apply for a green card one year after entering the country.⁴⁷ A green card confers legal permanent resident status.

How does the US compare with peers?

At a time when the number of refugees around the world is steadily rising, US resettlement rates are lagging.⁴⁸ Despite the growing number of refugees as counted under UNHCR’s mandate, the United States’ total resettlement fell short of Canada’s and Australia’s.⁴⁹ Even in absolute numbers, the United States fell behind Canada in 2018 and 2019, resettling 22,912 and 27,501 refugees to Canada’s 28,058 and 30,082 refugees.

On a per capita rate the United States’ refugee resettlement is also behind that of other countries. The US takes about 200,000 fewer refugees per year than it would be if the US rate matched Canada’s, for example. With a population more than double Canada’s, the United States can do much more. In fact, the US has done so in the past. The 1980 refugee cap was 231,700 and that year the US resettled 207,116 people.⁵⁰

Our analysis of UNHCR resettlement data, shown in figure 4, indicates that Canada resettled 800 refugees per million residents in 2019.⁵¹ Australia’s resettlement rate per capita was only slightly lower than Canada’s in the same year.

COVID-19’s effects are also evident in figure 4. Global refugee resettlement slowed in 2020 because of the pandemic.

When it comes to resettlement rate per capita, the United States is far behind. Though the US population is more than double Canada’s and Australia’s, US resettlement per capita has been much lower. In 2019, the United States resettled 84 refugees per million residents. In contrast, Canada resettled about 800 refugees per million residents that year. If the United States took in the same number of refugees per million residents as its northern neighbor, the United States would have resettled more than 258,000 refugees in 2019. That means even if the US hits the 125,000 refugee goal in 2023, it will be only half of the per capita level of Canada’s resettlement.⁵²

Ways for the US to increase resettlement to at least 250,000 a year

International comparisons make it clear that the United States could resettle many more refugees. Countries such as Canada and Germany take in many more refugees than the US. The US should set a goal to steadily increase the refugee cap over the next five years to levels comparable with international per capita levels. The United States should first aim to fulfill its 125,000 goal next year and then to reach 250,000 refugees a year by 2028. A potential timeline is shown in figure 5 where the US reaches 250,000 by resettling about 25,000 additional refugees each year from 2023 onward.

This increase is certainly substantial relative to the current refugee numbers. Yet even this substantial increase only raises US resettlement goals to comparable levels with countries such as Australia and Canada. In addition, it also raises the same level to the same numbers the US resettled in 1980 and 1981, though numbers dropped much lower even before 1985.⁵³

Famously, Germany, a country of just 81 million, took in 1 million Syrian refugees in 2015. This influx initially caused concerns about immigration to rise, for obvious reasons, but since then concerns have returned to pre-2015 levels despite continued resettlement and immigration.⁵⁴ Since the US population in 2015 was about 320 million, Germany’s refugee intake is akin to the US taking in around 4 million refugees during 2015. The US took in only about 70,000 refugees in 2015. The comparison with Germany is meant only to illustrate that increased resettlement in the US may raise similar short-term concerns—yet it is likely that they will fade away after only a few years, just as they did in Germany.

We will explore four promising options for expanding refugee resettlement:

Formalizing the way that US refugee admissions goals are created
Expanding the US private refugee resettlement capacity
Adopting state-based resettlement options
Expanding the refugee matching system already in use

All of these would expand refugee resettlement to varying degrees. This section provides an overview of these models and how they can be utilized to increase refugee resettlement. Whether it’s shifting resettlement’s costs to private sponsors or devolving more choice and authority to localities, these models can resolve some of the political tension around increasing refugee resettlement.

Each of the avenues for expanding refugee resettlement in the United States relies on allowing more choice in the refugee resettlement process. That is, they all devolve power from the federal government to others to choose the number of refugees and to select specific refugees or refugees with specific skills. In no proposal is that devolution absolute. States would still need to work within federal regulations or laws. Still, each avenue gives local groups, whether private organizations or state governments, more options to support resettlement without requiring anything from those who object to resettlement.

Formalizing the refugee resettlement goals process could make the resettlement system resilient and predictable

Today, the president is able to set the refugee admissions level after consulting with Congress. This generates large swings in refugee caps from president to president. The central challenge this creates is that it threatens the resettlement infrastructure. After a downturn it is difficult to build the programs back up.

Consider the Trump administration’s reduction in resettlement and the Biden administration’s difficulties in scaling back up. In 2017, under the Trump administration, the State Department concluded that if fewer refugees are coming to the US, then fewer resettlement offices will be needed. In 2017 the department informed the nine major refugee resettlement agencies that any office expecting to assist fewer than 100 refugees the following year would be closed.⁵⁵ This led to the closure of 134 offices, cutting resettlement capacity by 38 percent.⁵⁶ That is, one out of every three resettlement offices closed.

The fundamental cause of these closures is that offices receive funding on the basis of the number of refugees they are expected to resettle in a given year. So when numbers fell, so did funding. This certainly makes sense: there is no need to support an unused arm of any program.

However, a closed center is a problem because reopening, finding staff, and then training the staff all take time. People who left resettlement agencies are now working in other areas. These swings from low levels of resettlement to high-minded aspirations undermine the United States’ long-term refugee resettlement capacity. Today, despite the 2022 goal to resettle 125,000 refugees, the Biden administration is likely to resettle fewer than 30,000 refugees. A similar story played out during President Biden’s first year in office, when the goal was to resettle 62,500 people but fewer than 20,000 were admitted.⁵⁷

The Biden administration’s 2022 resettlement numbers exclude Afghans and Ukrainians because these groups were admitted largely outside the current refugee infrastructure. Despite the common conception (for good reason) that both groups are refugees, Afghans and Ukrainians were not admitted through the existing refugee system. Most were admitted under humanitarian parole. This means that they do not qualify for the supports available to refugees settled through the primary system.⁵⁸

Current US refugee resettlement capacity is so strained that regular refugee resettlement entirely stops when crises arise. For example, during the United States’ withdrawal from Afghanistan, the federal government put the ongoing Refugee Resettlement Program on hold for two and a half months, from October to January.⁵⁹

In a similar situation, during the first months of Russia’s invasion of Ukraine, immigration advocates pointed out that the US formally resettled almost no Ukrainian refugees. In March 2022, only 12 Ukrainian refugees were resettled through the refugee program.⁶⁰ This wasn’t because more were not coming to the US but because the current US system could not meet their needs. Still, significant numbers of parolees were entering from Afghanistan and Ukraine. About 72,500 Afghans and 39,000 Ukrainians were paroled into the country as of July 2022.⁶¹

There are two lessons to take away from these events. First, the current US refugee resettlement system is fragile and driven by political winds instead of the actual needs of refugees. It is difficult to rebuild the program, so lowering the cap does lasting damage. Second, simply raising the existing cap does not raise actual resettlement levels.

To address both of these problems, we propose a simple, needs-based formula to set a floor on refugee admissions.⁶² The first difference from current practice is that this would be a floor, not a ceiling. (The policy could allow for rolling unmet numbers into future years as well.) A second difference is that if the US were to set its floor to, say, take in 10 percent of the refugees in need of resettlement (RINOR) population,⁶³ this could replace the current system where the president consults Congress and then sets the refugee goal for the year.

A key advantage of setting a floor based on the overall refugee population is that this would not be a political choice. Instead, the floor would relate to the worldwide needs for resettlement. The US has agreed to play a part in helping refugees. A formulaic approach that centers on those most in need is a better way to fulfill obligations under this agreement.

In addition to being based on the global needs for resettlement, the formula could include measures of economic health, such as the unemployment rate or the number of jobs relative to the size of the US unemployed population. This would allow the US to draw down its refugee goals slightly in hard times but take more refugees in during economic expansions.

A needs-based formula to set a refugee resettlement ceiling

The RINOR category is a small portion of all the refugees in the world. Only those who are most in need of permanent resettlement are included in this measure. The current RINOR count is estimated at 1.47 million.⁶⁴

In line with a recommendation from the National Immigration Forum, a formula-based refugee floor could use 10 percent of the RINOR metric as a baseline.⁶⁵ This proportion could be set higher or lower depending on circumstances, but it would set a clear, predictable path for resettlement organizations to plan for. Again, measures of economic health can also be incorporated into the floor calculation, though simpler might be better for this proposal. To illustrate our proposal, figure 6 compares a hypothetical resettlement rate formula, which includes adjustments based on economic activity, to the historical refugee cap.

The UN began using RINOR in 2011, so our formulaic level starts the following year. The major factor driving a refugee cap based on this system is the global need for refugee resettlement. As the employment rate increases or decreases, the formula could automatically adjust so that the US takes in slightly less than 10 percent or slightly more than 10 percent of the year’s RINOR population.

For example, from 2014 to 2015, the employment rate increased from 68.14 percent to 68.70 percent. In response, the 2016 resettlement level could be set at a percentage of the 2015 RINOR level equal to 10 percent plus an additional 1 percentage point. So in 2016, the US would resettle 11 percent of the RINOR population. The RINOR level was about 958,000 in 2015, so the refugee cap for 2016 would have been about 105,000 refugees. Figure 6 shows the projections from this hypothetical from 2011 to 2023.

There’s no mathematical reason to merely add a percentage point to the RINOR resettlement goal when the employment rate grows. We do it only as an illustration of what could be done. The point is to make it simple and predictable, so nearly anything will do.

One benefit of a formula is that it makes debates about how many refugees to admit concrete. Opponents of resettlement could propose attaching or moderating the refugee admission level on the basis of the metrics that they care about—for example, public health, native-born births, economic health, or other metrics related to America’s ability to integrate migrants. But these concerns would also be weighed against the number of worldwide refugees most in need of resettlement.

Another advantage is that this formula will raise resettlement rates toward levels comparable to those of other countries. When we are deciding the right portion of the RINOR to bring into the US, we should keep an eye on international comparisons with similar countries. Most have much higher per capita rates of resettlement then the US. It would be simple to use international comparisons when selecting the portion of RINOR to use in our calculations. Assuming that 2024 looks like 2023, a 12 percent RINOR goal would mean that the US has a similar per capita resettlement rate as does Canada and meet our earlier suggestion of resettling about 250,000 per year.

Our proposal is not unlike others. The Guaranteed Refugee Admission Ceiling Enhancement (GRACE) Act is one example.⁶⁶ It would set a floor of 95,000 refugees resettled and require quarterly updates on the plans to reach the goal. Another is a 1997 proposal that emphasized the need to focus on the needs of refugees. The proposal called for sharing the “refugee burden” among stable and wealthier states.⁶⁷ If implementing a formula like the ones we have proposed is unappealing, then carrying out a reform like these others, which set floors and focus on those most in need rather than on erecting aspirational ceilings, could be a valuable step.

Creating certainty to improve resettlement outcomes

Fundamentally, no formula will eliminate all uncertainty, given the nature of refugee populations. Many important events are unpredictable—public health concerns loom large. Resettlement offices facing the uncertainty of what their resources will be during the next administration are cautious about expansion.⁶⁸ But creating certainty by changing the ceiling into a floor enables expansion. A floor set by a needs-based formula is more resilient than the United States’ current ceiling, which the country rarely reaches. Certainty will improve the United States’ ability to respond to international crises, like those created in Afghanistan, Ukraine, and elsewhere, by taking in more refugees through the usual refugee program.

Adding certainty to the resettlement process will enable consistent resettlement by agencies and avoid the problems that have previously caused resettlement to stop or slow.⁶⁹

Private refugee sponsorship increases resettlement without placing the burden on the federal government

One option for the United States to resettle additional refugees is to expand its new private sponsorship programs for refugees and establish additional sponsorship programs. The US Department of State announced the start of this program in the US on January 19, 2023.⁷⁰

Sponsorship is a tool that Canada currently uses. Canada has resettled more than 327,000 refugees since the beginning of its private sponsorship program in 1979.⁷¹ In 2019 alone, private sponsorships made up about 19,000 of 30,000 refugees resettled.⁷²

Private sponsorship allows private groups to sponsor refugees. In the Canadian sponsorship program, there are three ways to become a sponsor. The first is to be a sponsorship agreement holder (SAH). SAHs are typically religious, ethnic, community, or service organizations.⁷³ These sponsors have signed a formal agreement with the government of Canada to help support refugees. SAHs are prepared to sponsor an increasing number of refugees per year and typically have a large support network for regularly resettling refugees.⁷⁴ The second way to become a sponsor is to assemble a group of five or more Canadian citizens or permanent residents and submit an application for sponsorship together as a “group of five.”⁷⁵ The third way is designed for organizations that are not large enough to provide support as an SAH. These organizations, which include associations and corporations, can become “community sponsors.”⁷⁶

Private sponsorship reduces the costs of increased refugee resettlement for the federal government. Sponsors are responsible for providing refugees with financial support and integration assistance for up to one year after they arrive in Canada. The sponsor must have a specific amount of funds prepared to support the refugee. The Canadian government has cost estimates for these figures that it provides to interested sponsors.⁷⁷ Financial support includes initial resettlement costs for housing, ongoing monthly costs for basic necessities.⁷⁸ On average, the cost is $28,700 for a family of four.⁷⁹

Private sponsorship has become a major avenue for resettlement in Canada’s resettlement program as more and more private groups develop the capacity to resettle refugees. This approach increases the refugee cap without increasing the financial burden on the government and taxpayers.

The US is beginning to develop and roll out similar programs as Canada’s sponsorship system.⁸⁰ In a press statement in May 2021, US Secretary of State Antony Blinken announced that the US government has plans to launch private and community sponsorship programs.⁸¹ The details were released in January of 2023.⁸² They follow some of the Canadian system’s outline. Groups of at least five people can sponsor refugees by finding housing, picking them up at the airport, helping the refugees find work, and enrolling their children in school.

Before January 2023, sponsorship programs for Ukrainians had been announced and begun. In June 2022, news reports showed that at least 45,000 people in the US had applied to participate in these sponsorship programs. This was mainly in response to the invasion of Ukraine and the displacement of millions of Ukrainians.⁸³ This reactivity suggests that sponsorship could be effective at expanding US resettlement capabilities.

A similar system is beginning for Venezuelans as well. In October 2022, the Biden administration announced a sponsorship program for Venezuelans.⁸⁴ It allows 24,000 Venezuelans to be sponsored into the US. This program is meant to convert illegal border crossings into legally sponsored migrants.⁸⁵ As of November 3, 2022, about 28 percent of the 24,000 slots created in the program have been filled.⁸⁶ On the basis of the early popularity of this program, some immigration policy experts have called for expanding it once it reaches its current cap of 24,000.⁸⁷

A primary advantage of the private sponsorship program is that it ensures that the fiscal costs of refugee resettlement stay low. In effect, a sponsorship program allows people who want more resettlement in the country to prevent the costs of resettlement from being covered by others who do not want more refugees. This makes a private sponsorship program an avenue to increase the refugee cap without placing responsibility for increased resettlement on opponents of expanded immigration.

Policymakers’s concerns about private sponsorship options like those for Ukrainian refugees and Venezuelans is the long-term path that they create. Refugees through the normal process have a clear pathway to integration with American society because they have work authorization and can become citizens. Privately sponsored refugees should be given similar opportunities to work and become citizens.

Place-based refugee resettlement can expand the role of local and state governments

Place-based visas are another avenue to increasing the refugee cap that avoids political roadblocks from opponents of expanded resettlement programs.

In a place-based visa model, states grant visas to immigrant workers. Instead of tying visa holders to a single employer, these visas allow workers to move within their sponsoring state as their employment and opportunities change. The idea also mimics existing visa rules that limit where refugees can work. For example, temporary visa programs already include work limitations. H-1B visa holders’ current employers, for example, must formally submit a petition to permit them to move to new employers.⁸⁸ H-2A visas have similar requirements. Switching employers requires a formal petition to the US Citizenship and Immigration Services for H-2A visa holders.⁸⁹ As many have noted, a place-based visa program would be completely consistent with existing practice.⁹⁰

The only difference between current practice and a place-based system is that, instead of limiting immigrants to their current employer (as the existing visa system does), place-based visas require that the holders work in the area issuing the visa. There is no reason that states cannot create agreements to honor visas in a regional system, however. A place-based system is a good way to protect refugees from abuse by employers because it gives them more bargaining power and employment options than exist under current rules.

A place-based resettlement approach to increasing refugee admissions would alleviate opposition by allowing states to choose to resettle additional refugees. States wanting to utilize the refugee labor force will no longer be limited, while state policymakers opposed to resettlement can simply opt out.

Compare this approach to refugee resettlement to living in an apartment complex. When you rent an apartment, you have the right to have a certain number of roommates or even guests. Your neighbors cannot dictate whether you may have roommates and guests, but nor are they obligated to invite these roommates or guests into their apartments.⁹⁰This analogy is borrowed from Chandran Kukathas’s work on immigration. Chandran Kukathas, Immigration and Freedom (Princeton, NJ: Princeton University Press, 2021), 186.

Place-based visas would work in a similar way for refugee resettlement, allowing states to invite a certain number of additional refugees without burdening the neighboring states that oppose increased resettlement.

A state-based refugee resettlement process would allow the refugees admitted to work only in the state that admitted them. Travel would not be prohibited or affected by the visa. Only work authorization would be limited. That constraint mirrors the existing limits on temporary migrant workers, which tie them to a specific employer. The state-based work authorization could expire once a refugee obtains citizenship (which takes about five years) or on an accelerated path of one year once the refugee obtains a green card.⁹¹

Existing proposals for place-based visas focus on economic justifications, such as meeting industry needs or resolving labor shortages, but the program could be expanded to meet humanitarian needs as well. States could select refugees for both economic and humanitarian reasons depending on the program’s design. States like Utah, for example, could select refugees for their growing tourism sector, or they could select a number of refugees solely for humanitarian reasons.⁹²

A variety of place-based visa proposals exist. Most suggest a state-based geographical scale: the Texas Public Policy Foundation’s policy proposal, Canada’s Provincial Nominee Program, Australia’s Skilled Work Regional Visa program, and US Rep. John Curtis’s state-based visa pilot program (based on US Senator Ron Johnson’s 2017 bill).⁹³ Pete Buttigieg’s policy proposal and economists John Lettieri, Kenan Fikri, and Adam Ozimek’s “heartland” visa, on the other hand, outline a city- or community-based visa.⁹⁴ Their proposals specifically target places suffering from population stagnation and loss, aiming to lure business investment and expand the local tax base in declining cities.⁹⁵

Two challenges to place-based visa systems are questions about citizenship and questions about access to federal and state assistance programs.⁹⁶ The crux of both issues is that state-based refugee programs will eventually allow refugees to become citizens. Or refugee resettlement may draw on federal funds. Citizenship certainly makes new Americans eligible for federal programs. That eligibility means that even a state that opts out of resettlement could pay for some of a new citizen’s federally provided benefits. Given the positive findings that refugees pay more into public coffers than they remove, however, this is not a serious concern.⁹⁷

In net terms, states that opt out of additional resettlement will actually be free riders on the resettlement efforts of their neighbors. Refugees pay more into federal coffers than they take out, on average. Concern about costs could also be resolved by refunding states that don’t participate what they demonstrate is the costs that they face. Though the federal administrative costs implied by this are likely to be larger than the actual costs that place-based refugee sponsorship incurs for states that opt out of the program.

Another solution is again to devolve to the local level choices concerning state assistance programs, as already happens in states with refugee resettlement. Refugees resettled by means of a place-based system should follow the same citizenship and public assistance process as other refugees.

Place-based visa proposals are a template for expanding refugee programs. Devolving immigration choices to states enables states to decide the number and type of refugees to admit. Alternatively, they can opt out altogether, an option that should reduce resistance from opponents of refugee resettlement. To (again) compare immigration to living in an apartment complex, prohibiting refugee resettlement next door is like prohibiting neighboring apartments from hosting guests. There may be reasons to set some restrictions, such as noise level restrictions, but these should be targeted at the specific problems. Curtis’s proposal, for example, decreases the number of state-sponsored visas by 50 percent if more than 3 percent of visa holders violate the conditions of the visa in a year. Refugee policies can be designed similarly to target specific problems.

A refugee-serving, place-based visa program would allow states to opt in to an allotment of extra resettlement slots. For example, if the resettlement floor were set at 200,000, a portion of the visas could be established through a place-based visa program. Since this is a new proposal, it is best to give states wide authority to shape their own systems. Experimentation among the states is the best path toward finding what works best.

Refugee matching can meet global and local demands

Refugee matching is a system designed to match refugees to countries, local areas, or agencies.⁹⁸ Refugee preferences are matched to local preferences, local capacity, and resettlement agencies’ goals.⁹⁹ Refugee matching relies on the theory from the economics subfield of matching and sorting, but for a crude comparison, think of a dating service.

Matching systems give localities more control over the refugees they resettle, which encourages increased resettlement and brings in refugees who fit local needs.¹⁰⁰ For example, rural areas in need of home health aides and nurses would be able to select refugees with relevant training.

Most resettlement is done by hand today. Agencies look at the demographics of refugees and sort them into places that the agencies believe will most likely provide a successful match. For example, often they encourage refugees from the same country to group together so that those who have been in the area longer can help new arrivals. This system cannot meet the scale of the refugee challenge—there are more than 27 million refugees worldwide. It is too expensive for each refugee’s placement to be determined individually. In place of current methods, the US should fully adopt refugee-matching systems that it has run on a trial basis and should also experiment with additional matching systems.

This approach to resettlement also speeds up the resettlement process by using a computer algorithm to quickly determine the matching outcome. Both refugees and those resettling refugees submit their information and then the computer finds the best matches.¹⁰¹ In some ways, this is an automation of the existing processes. Resettlement agencies already sort refugees into areas where they expect them to be successful. This usually involves assessing the availability of community services, the likelihood of employment, and the presence of family members nearby. But a computer-based matching system can easily sort hundreds of thousands of people—or at least many more than current protocols, and at a lower cost.

Using automated matching introduces some constraints into the matching process. For example, it requires information on all the relevant factors that localities want to match refugees with. That is, you have to put in the information that a refugee has a medical background and the locality looking for refugees needs to note that they are looking for people with medical training. The trade-off is that including more factors can improve matches, but it also makes the calculation more difficult. The size of the resettlement country, for example, factors into the ranking system for refugee preferences.¹⁰² In larger countries like the United States, experts recommend ranking placement locations by type of area or by state rather than ranking smaller, individual localities, such as cities. Adding too many localities reduces accuracy in the matching system. The number of incoming refugees, the availability of housing, and service constraints (e.g., school seats, hospital beds, language classes, employment training programs) also add complexity to developing a matching system that can fully account for local and refugee preferences.¹⁰³

For policymakers seeking to implement a matching system, one relevant question is what information is available. There are at least seven different mechanisms for matching refugees with communities. Some are specifically tailored to the context of resettlement.¹⁰⁴ The choice of the specific matching algorithm depends on the information available about the preferences of refugees and about local priorities, as well as on the number of refugees, localities, and services and on the importance and heterogeneity of local priorities.¹⁰⁵

The ultimate results of matching systems are simple: lists of which refugees should go where.¹⁰⁶ In fact, US agencies are already using some matching systems to improve refugee resettlement in the country. A group of five economists created a matching system called Annie MOORE and helped a US refugee resettlement agency to implement the program. (The system is named after the first woman processed at Ellis Island.) “Annie” works by initially matching refugees with communities on the basis of the refugees’ characteristics and the preferences of the communities. After the initial match, the agency’s professionals can adjust the matches as they see fit.¹⁰⁷

Early research on the effects of Annie suggests that it can improve matching to the benefit of refugees and the US. At least in the short run, employment seemed to improve for refugees sorted by Annie. Considering that there are 27 million refugees in the world, manual evaluations and sorting are simply impossible. Annie’s creators conclude that Annie can reduce the costs of resettlement by requiring fewer personnel while simultaneously improving outcomes.¹⁰⁸

Improvements to the matching process are important both because of the scale of the world’s refugee challenge and because matching should improve outcomes. By sorting refugees and evaluating the match afterward, Annie’s creators expect the software to help refugees “more quickly integrate economically into each [resettlement] affiliate, as well as make more productive economic and societal contributions such as creating new jobs and generating tax revenues, [benefitting] local communities.”¹⁰⁹

Another, similar tool is GeoMatch. Created by Stanford University’s Immigration Policy Lab, GeoMatch allows localities and refugees to indicate their characteristics and preferences. Then it helps point immigrants toward the communities where they are most likely to be successful.¹¹⁰ The program is being tested in Canada and the Netherlands.¹¹¹ The work follows from similar matching programs, like Annie, but extends matching to apply to economic immigrants as well.¹¹² For example, Canada is matching skilled workers to areas in need of people with certain types of technical training.

With these tools, localities can take in more refugees knowing that these refugees will match their priorities. More control over decisions about which refugees to take may also encourage increased resettlement. Early results suggest that matching systems improve results over the current system for both refugees and receiving countries. Policymakers should accelerate their adoption in the US and internationally.

Conclusion

Raising US resettlement levels to be comparable with those of other countries will be a long-term project—one that will generate large and important benefits for the country and for the refugees themselves.

The US can both match and exceed other countries in refugee resettlement. Instead of adopting political goals that the US is unlikely to reach, like the 125,000 cap, policymakers should be changing the way refugees are admitted, which would allow us to set our sights higher—to at least 250,000. And we should be accepting only success in meeting our numerical goals. The proposals outlined in this paper are options for moving from here to there.

The US should not be content to sit on the sidelines as its peers act as a refuge for those in need. Our history of taking in the “wretched refuse” of the world benefits both the US and those walking through the golden door into their new home.¹¹³

Taking Politics Out of Refugee Policy: Improving US Resettlement for Those Most in Need

Executive Summary

Resources

Focuses

How does the US refugee system work?

Who is a refugee?

How many refugees exist in the world?

How do refugees get to the United States?

How are refugees resettled?

How does the US compare with peers?

Ways for the US to increase resettlement to at least 250,000 a year

Formalizing the refugee resettlement goals process could make the resettlement system resilient and predictable

A needs-based formula to set a refugee resettlement ceiling

Creating certainty to improve resettlement outcomes

Private refugee sponsorship increases resettlement without placing the burden on the federal government

Place-based refugee resettlement can expand the role of local and state governments

Refugee matching can meet global and local demands

Conclusion

Executive Summary

Resources

Related Publications

Related News

Related Blogs

Focuses

How does the US refugee system work?

Who is a refugee?

How many refugees exist in the world?

How do refugees get to the United States?

How are refugees resettled?

How does the US compare with peers?

Ways for the US to increase resettlement to at least 250,000 a year

Formalizing the refugee resettlement goals process could make the resettlement system resilient and predictable

A needs-based formula to set a refugee resettlement ceiling

Creating certainty to improve resettlement outcomes

Private refugee sponsorship increases resettlement without placing the burden on the federal government

Place-based refugee resettlement can expand the role of local and state governments

Refugee matching can meet global and local demands

Conclusion