Resettling refugees using preferences of refugees and hosts

 Here's the latest report from HIAS on matching Ukrainian refugees to hosts in the U.S.

How an Innovative Algorithm Helps Ukrainian Refugees Find New Homes  By Brian Zumhagen

"Odessa residents Max and Yuna* fled Ukraine on the day the Russian invasion began, February 24, 2022. It took them 7 days to reach the Polish border.

"The couple, both in their early 20s, spent the next several months in Poland. In September, they started applying for relocation to the United States with the help of HIAS. But unlike most refugees, Max and Yuna were among the first to use a new system that allowed them to list their preferences about where to be resettled, and any special needs they might have — thanks to a matching algorithm known as RUTH, which stands for Refugees Uniting Through HIAS. (The name was also inspired by the biblical Book of Ruth, which tells the story of how Ruth is herself welcomed as a foreigner).

...

"Back in Poland, HIAS Relocation Officer Denis Ruksha said some of the refugees from Ukraine he works with are relocated through European Welcome Circles, while others are resettled through circles in the United States. For those heading to the U.S., Ruksha has been using the RUTH platform for the last 3 months, entering beneficiaries’ preferences about where they would like to be relocated, along with other information. “It allows people to mention almost everything they think is relevant,” he said. In the U.S., volunteers in HIAS Welcome Circles can, in turn, enter their own preferences, such as the number of people they can host.

...

"RUTH isn’t the first computer system with a human name that HIAS has used to make its resettlement work easier and more effective. In 2018, the organization worked with partners to create matching software named after the first immigrant registered at Ellis Island in 1892. “Annie MOORE” (Matching and Outcome Optimization for Refugee Empowerment) used past employment data to direct refugees to locations where they would have the greatest chance of finding work.

"But where Annie focused on optimizing estimated employment outcomes, RUTH makes the relocation process faster and more transparent, according to the new platform’s developers. “This is the first time ever that preferences of refugees and priorities of hosts have been systematically used in the resettlement process,” said Andrew Trapp, associate professor of operations and industrial engineering at Worcester Polytechnic Institute.

"His colleague, Alexander Teytelboym, associate professor of economics at the University of Oxford, put it this way: “We think people are more likely to thrive in places where they prefer to live. Citizens are given a choice about almost anything of such consequence — so why shouldn’t refugees?”

********

Here are my previous posts on HIAS and refugee resettlement 

Matching refugees to homes: Oxford celebrates Alex Teytelboym's work with HIAS

Here's a timely short video from Oxford celebrating Alex Teytelboym's work with the venerable refugee resettlement organization HIAS. 

 

************

Earlier:  

Monday, April 29, 2019

Refugee resettlement in the U.S., by HIAS, using matching technology

Resettling refugees thoughtfully, by Trapp, Teytelboym, Martinello, Andersson, and Ahani

Here's a new paper, which emerges from a collaboration with my favorite refugee resettlement agency, the HIAS:

Placement Optimization in Refugee Resettlement
Andrew C. Trapp, Alexander Teytelboym, Alessandro Martinello,
Tommy Andersson, Narges Ahani

"This paper integrates machine learning and integer optimization technologies into the software Annie Moore (Matching and Outcome Optimization for Refugee Empowerment), named after Annie Moore, the first immigrant on record at Ellis Island, circa 1892. Annie is, to the best of our knowledge, the first software designed for resettlement agencies pre-arrival staff to recommend data driven, optimized matches between refugees and local affiliates while respecting refugee capacities. Annie was developed in close collaboration with representatives from all levels of Hebrew Immigrant Aid Society (HIAS), where a first version was deployed in May 2018. New features were regularly added until August 2018 when it was presented to the US State Department and all staff at HIAS."

***********

HT:  Tommy Andersson on the What are some dissertation-worthy topics in market design? thread at EconSpark

HIAS president Mark Hetfield on the refugee crisis, on NPR

HIAS President: U.S., Europe Treating Migrant Crisis Like 'Business As Usual'

The United States has taken in 1,500 Syrian refugees since the conflict in that country started four years ago. Our next guest thinks the U.S. could and should be doing a lot more. He's Mark Hetfield, president of HIAS - that used to be the Hebrew Immigrant Aid Society. It is the oldest voluntary resettlement agency in the world and one of the biggest in this country. Welcome to the program.

MARK HETFIELD: Thank you.

SIEGEL: The U.S. State Department has indicated the U.S. could accept as many 8,000 Syrians in the coming fiscal year. What more do you think the U.S. should be doing?

HETFIELD: The U.S. should be doing a lot more. We're living through the biggest refugee crisis, certainly, of my lifetime. We have 200,000 dead in Syria. We have people who are fleeing not once, but twice from the conflict. And frankly, the United States and many countries in Europe are treating this like it's business as usual. Taking 8,000 refugees - let alone the 1,800 they might take this year - is not a serious response.

SIEGEL: What's a serious response? What's a number that's a serious response?

HETFIELD: Frankly, well, look at what we did in 1980, during the Indochina boat crisis. We took it over 200,000 refugees with no infrastructure in place to do so, and these were boat people. These weren't people who were coming to our shores; they were in Asia. We were able to mobilize and take 200,000. We should be looking at that number today.

Refugee resettlement in the U.S., by HIAS, using matching technology

The Atlantic brings us up to date:
How Technology Could Revolutionize Refugee Resettlement
A software program called “Annie” uses machine learning to place refugees in cities where they are most likely to be welcomed and find success.

"Monken, an associate director at HIAS, a migrant-assistance charity, tells me Njabu and his family were specifically placed in Pittsburgh “because of the high employment probability forecasted by Annie.”

She was referring not to a person, but to a software program. Named for Annie Moore, the Irishwoman who was the first person to pass through Ellis Island, the New York outpost that served as the gateway for millions of immigrants to America, Annie is at the core of an ambitious experiment, one that, were it deployed more widely, could transform how refugees are allocated and treated around the world.
...
"Developed at Worcester Polytechnic Institute in Massachusetts, Lund University in Sweden, and the University of Oxford in Britain, the software uses what’s known as a matching algorithm to allocate refugees with no ties to the United States to their new homes. (Refugees with ties to the United States are resettled in places where they have family or community support; software isn’t involved in the process.)
...
"The software itself is in its infancy right now. For one, it is lacking in data: HIAS has been using Annie since last summer and has placed about 250 people via the software so far. There’s no exact number on how many refugees Annie must place in order to measure the program’s success. Instead, the software’s efficacy will be measured over several years and through the economic outcomes of the cases that go through the algorithm. Back-testing using data from previous years has yielded promising results, but the real outcomes will take a long time to discover. (Acquiring more data will be its own challenge: The Trump administration’s policy of reducing the number of refugees resettled in the United States means that last year the country accepted fewer refugees, just 22,491, than at any other point since President Jimmy Carter signed the Refugee Act of 1980.)"
************
Here's a website related to the algorithm and related academic work:

Refugees.AI
Resettlement that empowers refugees and communities

"Contact Us

If you would like to work with us on designing matching systems for refugee resettlement, please drop us a line."

************
Here's an earlier post on this work:
Thursday, October 4, 2018

Resettling refugees thoughtfully, by Trapp, Teytelboym, Martinello, Andersson, and Ahani

Refugee resettlement and the top trading cycles algorithm, by Farajzadeh, Killea, Teytelboym, and Trapp

 Here's a recent paper that (among other things) considers using the top trading cycles algorithm for matching refugees to sponsors (under a special program for Ukraine), to satisfy the location preferences of refugees.

Optimizing Sponsored Humanitarian Parole by Fatemeh Farajzadeh, Ryan B. Killea, Alexander Teytelboym, Andrew C. Trapp, working paper, 2023

Abstract: The United States has introduced a special humanitarian parole process for Ukrainian citizens in response to Russia’s 2022 invasion of Ukraine. To qualify for parole, Ukrainian applicants must have a sponsor in the United States. In collaboration with HIAS, a refugee resettlement agency involved in the parole process, we deployed RUTH (Refugees Uniting Through HIAS), a novel algorithmic matching system that is driven by the relocation preferences of refugees and the priorities of US sponsors. RUTH adapts Thakral [2016] Multiple-Waitlist Procedure (MWP) that combines the main First-In/First-Out (FIFO) queue with location specific FIFO queues in order to effectively manage the preferences of refugees and the supply of community sponsors. In addition to refugee preferences and sponsor priorities, RUTH incorporates various feasibility considerations such as community capacity, religious, and medical needs. The adapted mechanism is envy-free, efficient and strategy-proof for refugees. Our analysis reveals that refugee preferences over locations are diverse, even controlling for observables, by demonstrating the difficulty of solving a much simpler problem than modeling preferences directly from observables. We use our data for two counterfactual simulations. First, we consider the effects of increased waiting times for refugees on the quality of their matches. We find that with a periodic Top Trading Cycles algorithm, increasing period length from 24 days to 80 days, improves average rank of a refugee’s match from 3.20 to 2.44. On the other hand, using the available preference data RUTH achieved an average rank of 4.07 with a waiting time of 20 days. Second, we estimate the arrival rates of sponsors in each location that would be consistent with a long-run steady state. We find that more desirable locations (in terms of refugee preferences) require the highest arrival rates suggesting that preferences might be a useful indicator for investments in sponsorship capacity. Our study highlights the potential for preference-based algorithms such as RUTH to improve the efficiency and fairness of other rapidly-deployed humanitarian parole processes.

#######

Earlier:

Sunday, December 18, 2022

Resettling refugees using preferences of refugees and hosts

Migrants aren't widgets: refugee resettlement is a matching problem

Here's an op-ed published in Politico Europe today over my byline:

OPINION

Migrants aren’t widgets

An American Nobel economist’s pressing advice for Europe.

By ALVIN E. ROTH

 

9/3/15, 2:04 PM CET

The Mediterranean isn’t an effective barrier between Europe and refugee crises in Africa, Asia and the Middle East. Europe could turn this challenge into a manageable opportunity to protect refugee lives as well as its own economy. But European countries must first agree on a strategy recognizing that refugees are not widgets to be distributed or warehoused. They are people trying to make choices in their best interest. Those decisions are often a matter of life and death.

August began with news of Abdul Rahman Haroun, the Sudanese man who, after having already risked his life to reach Europe by boat, put his life in peril again, coming within yards of successfully crossing the “Chunnel” on foot to reach England and claim asylum before being arrested. Then, on August 27, 71 men, women and children, at least some of whom were Syrian, were found dead in a truck near Vienna. These refugees also had already somehow safely reached Europe, but boarded a smuggler’s truck to make it to another European destination. Instead, they suffocated and perished.

These stories are shocking but not surprising. The developing world hosts over 80 percent of asylum seekers, but a growing number are making their way to industrialized countries. These refugees are trying to get to specific countries within Europe. Sweden, for example, received 81,325 asylum seekers in 2014, or 8,365 refugees per one million Swedes. In contrast, while Greece had 34,422 boat arrivals in 2014, only 9,435 applied for asylum in Greece. That’s only 859 per one million Greeks.

Greece and Italy have weak economies and weaker asylum systems. So refugees continue to risk their lives and conceal their identities until arriving at a chosen destination, which means some European countries are getting far more refugee arrivals than others. In July, the Luxembourg presidency of the EU tried to get member states to pledge to relocate 40,000 refugees throughout Europe. While that would be a small fraction of asylum seekers, governments agreed to take in only 32,256 “redistributed” refugees.

Refugee relocation is what economists call a matching problem, in the sense that different refugees will thrive differently in different countries. Determining who should go where, and not just how many go to each country, should be a major goal of relocation policy.

Far more refugees will want to go to countries with thriving economies than those countries have been willing to take. But economic strength isn’t the only factor.

Mark Hetfield, CEO of HIAS, the refugee organization that helped resettle my wife, Emilie, when her family fled from Egypt, told me that “refugees go and integrate where they have family, where they have community, or where they think they can support themselves — in that order.”

The issue matters not just because we want refugees to do well, but because it’s hard to keep them where they don’t want to be. In the U.S., where refugees may relocate at will, this is especially clear.

As Hetfield says, “Many Somali refugees initially settled around the country subsequently migrated to Lewiston, Maine. Lewiston has a weak economy but an established Somali community. Consequently, efforts to resettle these refugees elsewhere in the U.S. were less effective than they could have been. Their preferences should have been taken into account from the start.”

In Europe, rules are less clear. Asylees in one EU country must often wait years before they can work in other EU countries. But even if they could effectively be prevented from relocating (and can they be?), preventing refugees from acting in their own best interest is not in Europe’s best interest either. Shouldn’t the goal be to integrate refugees into the economy where they can be the most productive? If that is a goal, the information and preferences of the refugees themselves about where they could thrive shouldn’t be ignored.

My wife, Emilie, is living proof that refugees, when resettled in a conducive environment, can be a boon to their adoptive countries. After getting her Ph.D. in cognitive psychology, Emilie helped pioneer a new discipline, cognitive engineering, focused on understanding the kinds of information that people need to solve the problems they face and making that information available to them. This work helps make many complicated systems safer. Others like her would gladly add their own contributions if given the chance.

We have learned in America that refugees can be assets, but that we could do a better job of integrating them into our economy. It will be hard for that to happen in Europe until there is a shared vision for an orderly process to allow refugees to go where they know they will do best.

Alvin Roth, who shared the 2012 Nobel Prize in Economics for his work on matching and market design, is a professor at Stanford University, and author of “Who Gets What and Why: The Hidden World of Matchmaking and Market Design” (HarperCollins, 2015).