From a flurry of hurricanes to the timing of the Jewish holidays, there was a lot to skew September U.S. hotel performance, which makes it difficult to draw conclusions on what it might mean going forward.
HENDERSONVILLE, Tennessee—What a month September was.
First Hurricane Harvey hit Texas and Louisiana in late August. Then Hurricane Irma barreled down on the Caribbean and Florida in early September. Weeks later, Hurricane Maria ravaged Puerto Rico.
Oh, and then the Jewish holiday shifted from October 2016 to September this year. In other words, the U.S. September data is extremely hard to read.
So proceed with caution as we lay out the top-line key performance indicators, keeping in mind that the impact of hurricanes is often a surge in demand (from displaced people) and a decrease in supply (from temporarily closed properties).
1. RevPAR streak continues
That all said, U.S. revenue per available room grew by 2.4%, driven up by a 1.4% increase in occupancy and 1% average-daily-rate growth. RevPAR has now increased in 91 consecutive months.
But let’s unpack this data just a little bit. As I said, the hurricane impact easily skews results up. So, total U.S. RevPAR for September was up 0.7% excluding Florida—where RevPAR grew 10%—and Texas (+18.3% RevPAR).
2. Record demand
In September, the industry sold 108 million roomnights, which is a record for September and roughly 3.5 million more than September 2016. This is the second strongest absolute demand increase this year; March added 4.5 million roomnights because of the shift in Easter to April. But when we exclude Texas and Florida, it becomes clear that those two states helped quite a bit.
In other words, the demand surge in two states from displaced residents, contractors, FEMA personnel and insurance adjusters almost doubled the demand increase for the U.S.
Here is some fun with numbers: Assuming that both states and the U.S. had seen September year-over-year demand grow at just the same rate as the August 2017 year-to-date demand increase, we would have expected the following:
So, in other words, it could be argued that all (and then some) of the incremental U.S. demand this month came from the two hurricane-impacted states.
3. Group occupancy down, transient up
Group occupancy declined sharply (-9.1%) and transient occupancy increased at a healthy clip (+2.3%). The Jewish holidays likely forced the move of group demand into the first three weeks and then into October. Hurricane-related demand is likely the underlying cause for the transient percent increase.
Obviously in October, the group data will look much better. By how much, you ask? Well, let’s look at the Easter shift impact on April and March to see if we can deduct any insights:
So, Easter moved from March 2016 to April 2017, and group occupancy swung wildly. It’s probably not too far-fetched to assume that the October group occupancy increase will be around 7% or so, given that group occupancy declined 9.1% this month.
4. Downward trend in average daily rate
Transient ADR declined for the first time since the recovery started (-1%), and the group ADR change was positive but only by a hair.
It’s a bit easier to see when you chart it.
Happy Halloween, I guess. This is scary stuff.
5. Pipeline data
The number of rooms in construction increased by 5.6% year over year to 188,500 rooms. This is a sequential decline of 0.5% from August, but that decline was not driven by rooms in construction in Texas or Florida that were taken out by the hurricane.
Rather, properties were opened—i.e. they moved out of the pipeline—and fewer projects started construction, hence the sequential decline in rooms. It is likely that the open date of some in-construction projects will be pushed back, but no projects were pushed back by a whole phase.
I showed this table last month and it is still enlightening in September:
So, the planning numbers have not really budged. That will drive slowing pipeline numbers going forward, which should bode well for the future of this industry, Airbnb growth notwithstanding.
This article represents an interpretation of data collected by STR, parent company of HNN. Please feel free to comment or contact an editor with any questions or concerns.