Looks scary to me. Photo by Stijn te Strake on Unsplash.

How Will Hurricane Irma Impact Puerto Rico’s Migration?

And, the real question, its finances!

Hurricane Irma is barrelling down on Puerto Rico. Here’s a terrifying map:

Big hurricane, small island.

So what’s going to happen to Puerto Rico? Well, lest you should think this is somehow normal and Puerto Rico is extremely well prepared…

We’re talking about a big problem here. Irma is Category 5. The last Category 5 Hurricane to hit Puerto Rico was Hurricane San Felipe II in 1928. In 1979 when Category 4 Hurricane David hit, and in 1989 category 4 Hugo hit. That 1928 hurricane killed over 300 Puerto Ricans and left half a million people homeless, and actually helped stimulate the wave of Puerto Rican migration to the United States. Throughout the Caribbean, Hurricane San Felipe II killed over 4,000 people, making it the 9th deadliest hurricane ever. It dumped between 2 and 2.5 feet of rain on the island, not quite the levels of Hurricane Harvey, but nothing to sneeze at.

Those more recent hurricanes like David, Hugo, and another, Georges, in 1995, also did enormous damage, despite being less powerful than Hurricane Irma. Take Georges, for instance. It left at least 20,000 people homeless, knocked out 96% of the electrical grid temporarily, and shut down water and sewage for 75% of the island. It was Category 3 with top wind speeds of 115 mph. Hurricane Irma right now is clocking in at 175 mph.

So we’re looking at a major disaster here. I feel for the people of Puerto Rico.

But this isn’t a meteorological blog. It’s a migration blog. Let’s talk about migration.


Puerto Rico’s Current Migration Status

I use airline embarkation data to keep track of Puerto Rico’s most up-to-date population movement trends. Here’s the latest data, which reflects domestic passenger balances through May of 2017, and international balances through February of 2017.

As you can see, Puerto Rico’s passenger balances have been improving throughout 2016 and early 2017. They’re showing some actual signs of demographic life here. As much as I’ve been a debbie downer about Puerto Rico’s demographic fortunes (worse than Ireland in the 1840s, etc), recent air passenger balances have substantially exceeded what I believed likely. My forecasts from May were substantially incorrect, although not to such a degree as to make a major shift in Puerto Rico’s long-run demographic fortunes.

What impact might Hurricane Irma have on these trends? Well, let’s look at a comparison. Hurricane Irene in 2011, which left nearly 1 million Puerto Ricans without power, dropped 22 inches of rain, and had windspeeds of up to 111 mph, is probably the most analog for Irma in our data. It was only category 1 or 2 when it hit Puerto Rico, but it may at least give us a rough indication of what we can expect. So what happened?

Well, I won’t bore you with the graphs, because the answer is nothing. That August saw a perfectly normal month of airplane traffic on both a gross and a net basis. September did too.

So what’s going on here? Well, when I say “perfectly normal,” I mean perfectly normal for 2011, that is, a normal month in a year that yielded about -40,000 for the whole year. So “normal” here still means “quite bad.” Maybe the other factors impacting Puerto Rico, like its recession, drove outmigration of people who would have been pushed away by Hurricane Irene. That’s possible.

But let’s look at some academic research.


Do Hurricanes Cause Migration?

I want to focus on two very recent papers. The first one is a working paper in this week’s NBER papers assessing what impact hurricanes have on migration to the US. That seems pretty fitting right now. So what does it say?

They find that hurricanes do cause immigration to the US, that this immigration is mostly legal immigration, and as such that it is highly influenced by the extent to which hurricane-impacted people can obtain legal status. The more easily impacted people can score legal immigration to the US, the more responsive flows to the US are to hurricane strength.

Guess what? Puerto Ricans have free movement to the US. Interestingly, the authors also find that temporary migration may be up to 50 times larger than permanent migration; i.e. people may immigrate to the US for a few months or years after a hurricane, then return. This seems especially plausible for Puerto Rico.

Unfortunately, the estimates produced in this paper aren’t structured in a way to easily allow me to pirate them for the grossly inappropriate use I would like, namely guesstimating a migratory impact of a specific hurricane (Irma) hitting a specific place (Puerto Rico). Prudent of the authors to refuse to give reckless readers such a tool, but nonetheless frustrating. This paper tells us that Puerto Rico’s migration to the US could reasonably be expected to rise, but I’m not sure whether it should rise by 100 people, 1,000 people, or 10,000 people.

Luckily, hurricanes are a popular topic. Back in May, another NBER paper looked at disaster migration in the US generally. This paper gives us really easy-to-recklessly-misuse effects of disasters on migration rates. For hurricanes generally, we get an effect that 1 additional hurricane lowers net migration by between 0.7 and 1.9 percentage points; and 1 super-severe disaster (100 deaths or more) lowers migration between 2.3 and 5.9 percentage points. Now, this data is all at the county level, and Puerto Rico is bigger than a county. Let’s say the effect size should be between one tenth and one twentieth as big for an area of Puerto Rico’s size. That means, depending on how bad this hurricane gets and how widespread the damage is, net migration should worsen between 0.04% and 0.59%. Let’s call it “0.05%” to “0.6%” for simplicity.

Well, if we presume that airline balances are a correct proxy for migration (Census Bureau doesn’t necessarily think so, but we’ll see who’s right when the 2020 Census rolls around), then 0.6 percentage points lower decadal migration for 1 additional disaster should mean somewhere between 2,000 and 22,000 lower net migration as a result of Hurricane Irma, between now and Census 2020. My bias would be towards the 2,000–8,000 end of that range.


What Will Puerto Rico Look Like in the Future?

This is all back-of-the-envelope stuff. We don’t yet know what impact Irma will have on Puerto Rico, the estimate range for a given physical impact’s effect on migration is large, and I’m extrapolating outside the sampling range of the studies I’m relying on. All extremely rough stuff.

But here’s the thing. Say Puerto Rico’s net migration falls from current Census-estimated -1.9% by 0.2 percentage points, easily in the range of possibility for a major hurricane, to -2.1% until 2020. Say from there it recovers by 0.1 percentage points per year, then flatlines once it reaches a balance. What will population look like?

Here, I’ll reproduce the graph I produced for my post in May comparing to Ireland in the 1840s, and showing some of my various Puerto Rico projections.

As you can see, this Irma-contingent scenario is worse than anything I’d forecast through 2020: I did not forecast that Puerto Rico might be slammed by the worst hurricane its faced since at least 1928. Mea culpa. And the effect is bad enough that even if you assume breakeven migration by 2040, which is not in any sense guaranteed, you’re looking at post-1840-Ireland levels of depopulation. Hurricane Irma might just be the kick in the pants Puerto Rico needs to really fall off of this demographic cliff into total, epochal-level demographic disaster.

Or, maybe not. Maybe the disaster will be well managed, the worst wind and rain will just miss the island, and the residents who haven’t left in the last 10 years will be those determined to stay no matter what. Maybe. Forecasting is always tricky. But it’s usually best not to rely on a forecast requiring numerous caveats, ifs, ands, or buts.


Conclusion

The same conclusion I always have when writing about Puerto Rico: pay attention, folks! We are watching a real live demographic and population collapse on a monumental scale and, outside of muni bond traders and Puerto Ricans, I don’t think the mainland economics and demographic community is really paying that much attention to what lessons can be learned. It’s time we should.

In the meantime, pray for the people of Puerto Rico, who may be in for a very rough time.

Check out my Podcast about the history of American migration.

If you like this post and want to see more research like it, I’d love for you to share it on Twitter or Facebook. Or, just as valuable for me, you can click the recommend button at the bottom of the page. Thanks!

Follow me on Twitter to keep up with what I’m writing and reading. Follow my Medium Collection at In a State of Migration if you want updates when I write new posts. And if you’re writing about migration too, feel free to submit a post to the collection!

I’m a native of Wilmore, Kentucky, a graduate of Transylvania University, and also the George Washington University’s Elliott School. My real job is as an economist at USDA’s Foreign Agricultural Service, where I analyze and forecast cotton market conditions. I’m married to a kickass Kentucky woman named Ruth.

My posts are not endorsed by and do not in any way represent the opinions of the United States government or any branch, department, agency, or division of it. My writing represents exclusively my own opinions. I did not receive any financial support or remuneration from any party for this research.