Many people have seen the overall TSA data showing how many people are starting their trips at US airports on the whole, but did you know that TSA also releases data by airport? This data is fascinating, and it’s a great window into exactly what is going on today on a more granular level.
Unfortunately — and probably unsurprisingly — TSA has clearly spent a great deal of time figuring out how to release the data in the most unfriendly format possible. If you’re curious, you can find the data in a miserable PDF file with hundreds and hundreds of pages released on a weekly basis. The data is by checkpoint, by hour, and by airport, so there is no way to make this usable without a lot of manipulation. Fortunately, Seth Miller over at PaxEx.Aero did the hard work and kindly shared the data with me.
When I did this analysis, the most recent week available was June 28 through July 4. So, I took that and compared it to the same week the year before. Of course there are issues with holiday timing, but ignore that. I’m less concerned about exact change in numbers year-over-year and more interested in the variance of one airport versus another.
I narrowed the field by focusing only on the 145 airports that had at least 10,000 people go through TSA checkpoints during that week in 2019. For small airports, percentage changes tend to look wonky. Here’s the distribution showing the percentage of passengers passing through checkpoints this year versus last.
This is probably about what you’d expect. The bulk of the airports saw 2020 screenings drop to between 20 and 40 percent of the 2019 numbers. Of course, it’s the outliers that are more interesting.
On the very bottom end, it tells us very little. Those two airports were Saipan and Guam which had just about no service on purpose. Into the next bracket, however, it was mostly airports in Hawai’i with some big mainland airports as well. That makes for strange bedfellows.
The Hawaiian airports make sense. After all, Hawai’i is virtually shut down, so you’d expect to see far fewer travelers than normal. But here you also find New York/JFK, Washington/Dulles, San Francisco, Miami, Los Angeles… you get the point. What’s going on here is that the airports that rely on international flying more than most have a lot fewer people traveling than last year.
I took a look at 2019 numbers and pulled out the airports that have more than 15 percent of departures going internationally. International travel is much softer than domestic, and these numbers show that very clearly.
|Airport||% Intl Flights||TSA % of Last Year|
JFK was hit very hard by this, because travel over the Atlantic basically disappeared. Then Newark had the one-two punch of that and a very conservative United to knock it down. In fact, you see a whole lot of United hubs on this list.
Unsurprisingly, the only airport that cracked 20 percent of last year’s numbers was Fort Lauderdale. That airport may have a lot of international travel, but it also has a ton of low-cost carrier leisure flying. It stands to reason that it would see more people traveling than Miami for that reason alone. People were filing up Spirit airplanes.
At the other end of the spectrum, there are 6 airports with more than 50 percent of boardings year-over-year. They were all in Florida or Arizona. (Destin, Key West, Panama City, St Petersburg, Phoenix/Mesa, Punta Gorda) What do those have in common? Well, those last three are big Allegiant bases with little or no service by anyone else. It just shows that the heavy leisure focus has brought people back sooner… (or did until things started turning down again).
If you break out Arizona, Florida, and Texas, you see a different story compared to the other states.
Traffic had indeed rebounded much quicker in those states. Just seeing these big difference had me captivated. And before I finish up, I want to do one more thing.
Let’s take a look at all airports in California that hit the 10,000 boarding threshold. This is not the easiest chart to digest, so you may want to sit with it a bit.
Focus on the dark orange bars. Those are the TSA year-over-year screening percentages that correlate with everything else we’ve been talking about in this post. It is sorted from lowest to highest. The light orange is the percent of seats offered YOY in the market. In purple, you’ll find the percent operated by the Big Three airlines. Then the gray line at top shows the percent of seats that are domestic at each airport.
With this, we can figure out a few things. At the bottom end, LA and San Francisco are unsurprisingly low because they have such high percentages of international travel. Meanwhile, Long Beach sits at the bottom probably because of the JetBlue saga plus… Southwest.
Look at San Jose and Santa Barbara. They have similar TSA numbers YOY, but Santa Barbara fills those screening lines with far fewer available seats for sale compared to the year before. This looks like a Southwest issue, and you’ll see the same thing with Oakland and Burbank. Southwest has been running more flights but it’s been filling airplanes less.
Now look at the top four airports. What do they have in common? They’re all further inland. The inland parts of California behave more like the middle of the country than the coasts when it comes to willingness to travel and initial impact from the virus.
For as much as we can figure out here, there are still some outliers. Look at the top and you’ll see Ontario. You’d think it would behave similarly to Sacramento since it has similar airline profiles and geography. BUT, Ontario is part of the LA area, so airlines can pull out now if they want according to CARES Act rules. That’s not an option in Sacramento. Yet Ontario is still holding up quite well in comparison to others.
I suppose that perspective is the real story here. If Ontario is considered to be crushing it with just over 40 percent of last year’s screenings, that shows just how bad the industry is doing. And this was all happening before things started to saga once again.
Thanks again to Seth at PaxEx.Aero for sharing the data in a usable format.