Drought and Mexican migration

National Geographic News contacted me last week to ask for comments on a new study that suggests Mexican migration to the US rises whenever there's a drought in Mexico. It's an interesting study; the first of its kind so far as I'm aware. The authors based their findings on a statistical analysis of Mexico-US migration from 1995-2005 and compared it with climatic trends and crop yields in Mexico over the same time. The study's authors found that whenever crop yields went down significantly, migration to the US rose. The authors then go on to use their findings to predict how climate change might affect future Mexico-US migration. Taking model projections of future climate in Mexico, the authors suggest that an extra 1.4-6.7 million Mexicans might migrate to the US by 2080 because of climate-related crop losses. Here's the link to the original study, which was published in Proceedings of the National Academy of Sciences.

This study is one of the first I've seen that uses known migration and climate data to try and estimate the effect climatic events have on ongoing migration movements. There are studies that have estimated the number of people displaced by famines in East Africa, but that's a different scenario: in a famine, where you have people streaming out of the countryside to relief centres, it's relatively straightforward to make estimates of the number of people affected. But there was no famine in Mexico in the past 15 years; the countryside remained populated, relief centres were not established. Rather, what the authors were able to do was show that the regular and ongoing flow of cross-border migration responds to climate-related changes in crop yields.

This is rather important. Until now, most climate migration literature has focused on the 'environmental refugee' scenario of starving or flooded out millions fleeing areas rendered uninhabitable by extreme weather events, rising sea levels and the like. Many estimates are bandied about regarding how many climate change refugees we might expect in coming decades. The numbers range from tens of millions to a billion depending on the study you read. But most of those studies are back-of-the-envelope/best guess calculations.

Most migration worldwide does not occur as distress migration. People's motives for migrating vary considerably from one place to the next and from one household to the next. Climatic variables can and sometimes do play a role in household decision-making, but typically in concert with other social, economic and cultural influences. As a result, the greatest effects of climate change on human migration patterns are likely to be seen not in terms of the number of environmental refugees but on voluntary migration flows. Yes, environmental displacements will continue to occur; a surge in illegal migration by Hondurans to the US occurred in the months following 1999's Hurricane Mitch, for example. But the bigger numbers will be experienced as an increase in rates of rural-to-urban migration in developing nations - something that's very difficult to tease out empirically, since there are a lot of other factors also driving rural-to-urban migration.

A final observation is that while I'm pleased the authors were able to detect an association between crop failures and Mexico-US migration, I wish they had left it at that. There is so much inherent uncertainty in future climate models and in current migration data, when you add to it our complete inability to know future political, social and economic conditions in Mexico and the US, trying to predict climate-related migration over the next 70 years is a bit of a folly. Especially when the results come with falsely precise numbers like 1.4 and 6.7. Why not at least follow convention and round them, or simply state an order of magnitude (e.g. the extra migration is likely to number in the millions over coming decades)? Making predictions like this simply provides fuel to those who simply refuse to believe the science (see some of the comments on the National Geographic story to see what I mean) .