June 5, 2024

Moral hazard and migration

Vikram Maheshri discusses unintended consequences of search and rescue operations in the Mediterranean Sea.

Refugees make the voyage from Turkey to Greece on an inflatable raft.

Source: Anjo Kan

Since 2014, over 15,000 migrants have died or gone missing trying to make the voyage from the north coast of Africa to southern Europe. In response, European authorities have launched several search and rescue operations. There are few signs that migration along this deadly route is slowing down. In fact, efforts to curb migrant deaths may encourage even more migrants to make the perilous journey.

In a paper in the American Economic Journal: Economic Policy, authors Claudio Deiana, Vikram Maheshri, and Giovanni Mastrobuoni found evidence that migrants and smugglers responded to search and rescue operations by attempting even more dangerous crossings. However, the authors still say that such operations are likely beneficial to migrants on the whole.

Maheshri recently spoke with Tyler Smith about the impact of search and rescue operations on the market for smuggling along the Central Mediterranean Route and what policymakers should do to reduce migrant deaths.

The edited highlights of that conversation are below, and the full interview can be heard using the podcast player.


Tyler Smith: What's the relevant background for understanding your paper?

Vikram Maheshri: The 2010s were a period of tremendous upheaval in the Middle East and North Africa. The Arab Spring and other kinds of revolutions following that led to a lot of migrants trying to escape North Africa and the Middle East and make their way into Europe seeking asylum. One of the primary routes was through the central Mediterranean. If you look at a map, Tunisia and Libya are really close to Italy, so this was an attractive way to try to escape these awful conditions. In response to shipwrecks that killed numerous people, various bodies in Europe considered sending out big search and rescue operations, and these were intended to help these migrants at sea.

The debate about this was pretty simplistic. You had one group of people saying that they didn't want these search and rescue operations—that they didn’t want to get more migrants; that it was not good for Europe. And then you had another group of people saying that criticizing search and rescue operations meant you were against migrants and you're trying to kill them. That's an oversimplification of the debate on it, but maybe not an oversimplification of the loudest voices. In that context, we were wondering what's actually going on when these search and rescue operations are implemented. What happened to this market for human smuggling across the Mediterranean? Is it actually inducing more people to attempt this crossing? Is it making it less safe? Are there benefits accruing to migrants? Are there benefits accruing to smugglers?

Smith: What are some of the pieces of this market to keep in mind? For instance, what do migrants need to do in order to procure these services? 

Maheshri: Most of the people who attempted this crossing were coming from sub-Saharan Africa. So the first part of their journey would be this arduous trek across the Sahara, and then they often wind up on the coast of Libya. In these areas, the UN estimated at some point that there were about half a million people that were ready to cross at any given point in time. These people would make contact with smugglers, often through Facebook where posts in Arabic would say where to go and how much money it would cost. The cost of this varied between about $700 to $2,500, depending on the type of boat you were taken over on. Most migrants were trying, upon reaching Italy or possibly Malta, to enter the asylum system.

Smith: Can you describe these search and rescue operations a little bit?

Maheshri: It is a key part of our paper that there was a lot of variation in these search and rescue operations. Some of these operations were patrolling only within, roughly 30 to 50 miles of the European coast, whereas others would patrol as close to approximately 12 miles off of the coast of Libya, so very close. The patrol areas varied throughout this period. There were a number of specific operations that were changing over time. In periods of very intense search and rescue operations, the strategy of a smuggler would be to load up a bunch of people on a boat, take them out to sea and into the range of search and rescue, then throw a satellite phone in the boat with a number to call the authorities to try to get picked up. It’s a pretty harrowing process.

Smith: Obviously this is not legal migration, which I'd imagine makes this a really difficult issue to study quantitatively. What challenges did you run into when trying to understand the impact of these search and rescue operations?

Maheshri: It's a data-poor environment when you're dealing with illegal activities. It's often not monitored well. So we were able to get information on daily arrivals to Italian shores from the Italian state police. We didn't know exactly when they left, but we knew when they reached the shores. We were also able to get data on deaths in transit through three or four different organizations. And then the other challenge is that these operations weren't varying very rapidly over time, so we didn't have these very rapid natural experiments where we could compare what the world was like before and after an event. We also didn't have variation in one part of the Mediterranean versus another; search and rescue operations covered the entire central Mediterranean. So those were the challenges that we faced. But we did have one other piece of information. We had information on crossing conditions. There was a whole bunch of meteorological and tidal data out there, and we could figure out basically how rough the journey was.

What we really need to do is give these migrants an alternative to using these smugglers, and there's two obvious alternatives. The first alternative is to make it better in their home country, but that's really hard. The other possibility is to increase legal alternatives to the system of migration, but that is politically very fraught.

Vikram Maheshri

Smith: How did that information help you to answer this question about the impact of the search and rescue operations?

Maheshri: We could see how people responded to changing crossing conditions when the waves were really bad, when the weather was really rough. And with the aid of a pretty simple model—some pretty basic assumptions on how boats might cross the ocean—we were able to conclude a few things. One of these things was that if people responded to crossing conditions differently when there was an operation in place from when there wasn't an operation in place, then that would also imply that the total number of crossing attempts would also change in a predictable way. If people were less responsive to the bad crossing conditions when there was an operation in place, that would also imply that an operation would generally increase the number of people attempting to cross. It would induce some demand.

Smith: When you put the data that you have and this simple model together, what did you find?

Maheshri: The first thing we found is that crossing attempts are elastic to crossing conditions, so people do respond. They're less likely to attempt to cross on bad days. However, they're way less responsive when the search and rescue operations are in place. What that suggests is that these search and rescue operations are changing the decision-making of migrants. The second piece of information is that smugglers also have a decision to make here, because they can try to send migrants on sturdier wooden boats or they could try to send them on inflatable rafts, which are much more risky. However, they're also dramatically cheaper, both to the smuggler and to the migrant. What we find is that there's a huge shift by smugglers towards the use of these flimsy inflatable boats. And that does two things. First, you get more people attempting to cross. The reason is it's much cheaper to cross. The second thing is it's not obvious whether it's going to become more risky or less risky, because on the one hand, the boats are flimsier, but on the other hand, you have more people out there that are trying to help rescue. Ultimately, we find that it almost certainly became more risky on average to attempt to cross. That doesn't necessarily mean migrants are worse off, because a lot of migrants are still better off attempting this risky journey at a lower price. On average, migrant welfare probably increased. However, these benefits from the search and rescue operations were shared between migrants and smugglers.

Smith: Do you think that this research offers any lessons for policymakers in Europe? What should they attempt to do in order to reduce these migrant deaths?

Maheshri: I think that there are really two things that you can do. What we really need to do is give these migrants an alternative to using these smugglers. And there's two obvious alternatives. The first alternative is to make it better in their home country. But that's really hard. If we knew how to do that, there would be a lot of development economists looking for new work. The other possibility, of course, is to increase legal alternatives to the system of migration. And that is politically very fraught. But if you could find other means, that would be much safer.

Migrants at Sea: Unintended Consequences of Search and Rescue Operations” appears in the May 2024 issue of the American Economic Journal: Economic Policy. Music in the audio is by Podington Bear.