HENDERSONVILLE, Tennessee—I regret to inform you of the untimely demise of the U.S. hotel industry revenue-per-available-room up cycle.
After a short but valiant fight against last year’s hurricane comps, the demand increase from last year was just too much to overcome. RevPAR growth will be missed. It leaves behind a legacy of demand records, a mature and well-established pipeline and immature room rate growth that never really grew up. In lieu of flowers please attend next year’s Americas Lodging Investment Summit and celebrate the new RevPAR up cycle, which should by then be three months old.
1. RevPAR dip headlines KPIs
RevPAR declined 0.3%, driven down by occupancy declines of 2.1% and only partially salvaged by an average-daily-rate increase of 1.9%. In September, 107.5 million roomnights were sold—160,000 fewer rooms than a year ago—meaning that for the first time in 37 months, room demand declined. Supply, of course, increased 2%. So after three months of 70%+ occupancy, utilization declined to 68%. ADR growth was the lowest it has been this year.
So, is this The Big One? Is this the beginning of the end of the up cycle? Well, if you beat up the data long enough, it will confess to anything, even underlying strength. So let’s just exclude one market—Houston—from the U.S. data and, ta da … all is well again.
Or, to quote Mark Twain: The reports of my death have been greatly exaggerated.
2. Segmentation data
The top U.S. RevPAR line was down, but for the upper-end hotels, segmentation demand was actually positive. Transient rooms sold increased 0.1% and group room demand was up 1.2%. But because of the steady supply growth, occupancy declined (transient: -2.3%, group: -1.1%). These declines needed strong ADR growth to counter them but only group ADR increased (+3.6%) at a pace that helped lift RevPAR (+2.5%). Transient ADR (+2.2%) lead to flat (-0.1%) RevPAR change. This is the first time this year—and since March 2017—that transient RevPAR declined.
Now, let’s see if we can do our little magic trick again and exclude Houston from the U.S. dollar data with positive results:
Ah well, it was worth a shot. The point is that a lot of displaced folks in Houston last year were classified as “group” and the transient data did not really move the needle, so even after excluding Houston the transient data is still weak.
But it turns out that people who fled the weather made it as far north as Atlanta, pushing up demand there as well. So, let’s do the same exercise:
I just wanted to point out that when you exclude Atlanta from the segment data or Houston from the total U.S. data, your results are quite different and still positive. This points at the underlying strength of U.S. room demand.
3. The hurricane effect on the top 25 markets
The largest markets recorded a RevPAR decline of 1.4%, because of a 1.4% demand decline and a 2.6% supply increase. At the same time, ADR actually increased 2.6%. The takeaway is that the total U.S. and the top markets were dragged down by Houston and Orlando but buoyed by some other markets (see below), and the top line hides the actual positive monthly numbers.
Nine of the top 25 markets registered demand declines, but because of strong new supply, 18 markets saw occupancy decline. What’s interesting to overlay here that in only eight markets ADR decreased. A possible reason for that is the average occupancy in the markets was 73.2%, and in 17 markets hoteliers sold over seven out of ten rooms each night. That should give you revenue managers a sense of stability and hence pricing power.
Here are the top and bottom five markets when sorted by RevPAR % change:
The hurricane impact is clear. Displaced people went to hotels in Dallas and Atlanta and stayed in hotels near their affected homes in Tampa, Orlando and Houston. But Houston, wow—demand declined almost 30%, and in Orlando it was a decline of just under 10%.
San Francisco had a monster month, so it’s probably time to pay a bit more attention to its meeting market, now that we are in meeting season and the Moscone Center construction is nearing full completion in early 2019. Oh look, good results!
4. Pipeline data and more rooms in planning
Another month, another flat change in the number of rooms in construction. The total declined 0.2%, so now we have registered a decline in nine of the last ten months. Overbuilding won’t be an issue anytime soon.
The increase in the number of hotel rooms in the planning stage is on par with the fluctuations we have observed.
Digging a bit deeper, you can see that the midscale planning phase increased year over year by 131%, from 19,000 to 44,000 rooms. Digging more into that 25,000-room increase reveals that 21,000 rooms are under flags that did not really exist a year ago: Tru by Hilton (9,500 rooms) and Avid by IHG (11,600 rooms). So, the midscale supply increases will continue but that does not really pose a threat to the U.S. data, in my opinion.
5. Comments about Q3 2018
Not surprisingly, the monthly data dragged the rest of the quarter so RevPAR increased only 1.7%, the lowest monthly performance this up cycle (starting Q2 2010).
What is a bit more concerning given the low ADR growth is the “real” ADR chart showing ADR minus CPI, which we use as an indicator of the rate of inflation.
The real ADR percent change was -0.5%. So, in other words, the industry is not really in step with expense growth and that hurts profits.
Chains and independents
I learned about a news GIS software and had some fun with it. I uploaded every hotel in our global census database with an open date, back to the 1970s, and coded it by “chain” vs “independent.” Then the program gives you this neat time lapse of all hotel openings, ever, around the world.
It’s interesting how in the most recent past the majority of new hotels globally—not just U.S.—are chain-affiliated. (Click on “Total” in the upper left hand to see the running total). Here is a screen shot:
Jan Freitag is the SVP of lodging insights at STR.
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.
