BROOMFIELD, Colorado—The U.S. hotel industry hit the elusive 65% occupancy mark in April 2015, and the closest it had ever come before was back in the expansion period of the mid-1990s.
Since the occupancy record was broken last year, it has inched even higher, and as of January 2016 sat at 65.5%. But as occupancy levels continue at record levels, as new supply remains relatively low and even as inflation-adjusted average rates have surpassed their previous peak level, concerns still remain over a perceived lack of sufficient rate growth.
It essentially boils down to this question: At record occupancy levels, what should average rate growth be?
There's no easy way to answer this question, but it's worth exploring. Numerous factors—tangible and otherwise—play into the complicated business of hotel pricing, but does roughly 4% rate growth reflect an appropriate level during a time of record occupancy?
Comparing rate growth in 2015 and 1996
One way to examine rate growth is to see what happened in other cycles. Back in 1996, occupancy levels were similar to where they are now, new supply was relatively low but creeping up and existing supply was about 70% of what's in the market today.
STR, parent company of Hotel News Now, analyzed 27,561 properties reporting data consistently from January 2013 through December 2015, and 10,015 properties reporting data to STR consistently from January 1994 through December 1996.
It's important to keep in mind this is a same-store analysis; all properties had to be open and reporting consistently for the three-year period to be considered. As such, the overall samples are lower than the entire U.S. sample reporting to STR, and the inflated rate-growth effect of new properties entering the market is therefore removed. The performance comparison of the two samples is detailed below.
Both samples ran similar occupancy levels, which—when only considering hotels that were open and consistently reporting—were about three occupancy points above the approximate 65% level for all total reporting hotels.
In the 2015 sample, occupancy growth was minimal and rate growth was 3.9%. In contrast, in 1996, occupancy level was slightly down from the prior year, yet rate growth was a healthy 7.3%. Even when accounting for the inflationary differences between the two time periods, the 1996 sample achieved rate growth approximately 150 basis points above the 2015 sample. Using this general analysis, rate growth levels in 2015 were unable to achieve the growth demonstrated at a similar historical point in the industry.
We can take this a step further and even look at a true same-store analysis, and isolate just those properties that were open and operating in both time periods. This sample size is obviously smaller, but is nonetheless large enough to warrant analysis.
In this instance, there were more than 6,800 hotels that reported in both the mid-1990s and the mid-2010s. The rate growth results are nearly identical when considering the full sample and same-store sample, and the rate growth spread from the 1996 results to the 2015 results was exactly the same. Thus, we are left with the same answer: Hotels in 1996 had greater success with growing rate than hotels in 2015.
Busy hotels unable to capitalize on rate growth
If the two samples of hotels running 68% occupancy grew average rate at a different pace, what about hotels running at 80% or higher occupancy? The tenets of supply and demand would suggest those properties should be achieving even greater rate gains because they are effectively sold out many days of the week. Were higher-occupancy hotels able to drive rate more than other hotels, and were there still differences in rate growth between 1996 and 2015?
We split the previous samples of hotels from 1996 and 2015 into two groups: those running at 80% or higher occupancy and those running under 80%. We mandated 80% occupancy for two full, consecutive years (1994-1995, 2014-2015) so that these resulting hotels weren't varying significantly in occupancy for the measured periods.
Here is where things really get interesting. First of all, the results from the previous analysis were not only confirmed in this new
|
analysis, but the rate spread widened. Hotels running 80% or greater occupancy in 1996 grew rate 400 basis points higher than hotels running 80% occupancy or more in 2015. That same spread was 330 points for hotels running under 80% occupancy.
For all the hotels running above 80% occupancy, rate growth was stronger than those hotels under 80%, which makes sense. But in 2015, this spread was fairly minimal—only 40 basis points—meaning hotels that were running 86% occupancy in 2015 were growing rate less than half a point stronger than those hotels running 65% occupancy. In 1996, this spread was greater, about 110 basis points.
Perhaps the most interesting takeaway from the above chart is that, in 2015, hotels running at 80% or higher in occupancy actually experienced less revenue-per-available-room gain than those hotels running under 80%. Clearly, there's not as much room to grow in terms of occupancy for the busier hotels, but they seemed to be unable to capitalize on their position by maximizing rate growth. In 1996, the busier hotels had the greater RevPAR growth.
The New York City influence
Once we looked deeper into the distribution of hotels in this analysis, it was clear that hotels in the New York City market were influencing the 2015 results.
Specifically, when we looked at the hotels running 80% occupancy or greater, New York City hotels accounted for 20.1% of the rooms revenue for the 2015 sample, compared to 12.3% of the 1996 sample. And because New York City hotels saw rate declines in 2015, that had an effect on the overall numbers.
If we removed the New York City market altogether from each sample, the overall margin on rate growth narrows by about 80 basis points (though, in that case, hotels in 1996 still experienced greater rate growth than hotels in 2015, even after adjusting for the inflation rate). However, clearly, New York City hotels explain some of the reason behind the differences in rate growth.
The rate growth difference in the New York City market was massive between the two periods, a full 15-point swing. Much has been made about the high-occupancy/declining rate of New York hotels in 2015, with no clear answers. Popular theories include:
- the strength of the dollar in 2015, making international travel to the U.S. prohibitively expensive;
- huge influx of new supply to the market, accounting for direct and indirect pricing pressures;
- proliferation of alternative accommodations, such as Airbnb, which competes at a lower price point than hotels in New York; and
- the market hosted the Super Bowl in 2014, so rates in February 2015 were much lower relative to the prior year.
In terms of the Super Bowl, it’s true that average daily rate was down the following January (-8.7%) and February (-4.6%), but rate performance was also down five other months of 2015, and down again another 4.4% in January 2016. Thus, it's unlikely that the absence of the Super Bowl played a major role in rate declines in New York in 2015.
Supply is a real concern, and that seems to be a valid reason behind rate declines. But then again, we are still talking about hotels averaging just less than 90% occupancy. If supply was the real issue, would occupancy be so high? Or are operators anticipating the impact of supply and adjusting their rates ahead of the brunt of openings?
As for the respective impacts of currency exchange and alternative accommodations, both of these are valid issues worth noting. If these are the reasons for 2015 pricing weakness in New York, it's also reasonable to assume other major gateway cities would show similar trends. In our 2015 set of hotels running above 80% occupancy, the market with the second greatest rooms revenue contribution was San Francisco/San Mateo, a location that is a top destination for international travelers, and also has an ample supply of alternative accommodations.
For the San Francisco/San Mateo market, rate growth was much stronger in 1996 than in 2015, but in each case growth was above the pace demonstrated by the sample as a whole. This was not the case for the New York market, as shown below.
In other words, in both 1996 and 2015, busy San Francisco/San Mateo hotels grew rate anywhere from 38% to 56% above the average demonstrated by all national hotels over 80% occupancy. In New York, while hotels grew rate more than 50% higher than the average in 1996, they declined by more than 50% of the average in 2015. So while the issues of exchange rate and alternative accommodations might be playing a part here, there must be other factors accounting for New York's rate declines in 2015, such as supply concerns, which isn't an issue in San Francisco/San Mateo.
So what is it?
While New York accounts for a significant part of softer 2015 rate growth, the fact remains that even when excluding New York, hotels had less rate growth in 2015 than in 1996, even at similar occupancy levels. It would be too simplistic to come up with one magic answer to explain the difference, for there are so many moving pieces to account for (total amount of hotel supply, consumer confidence, stock market movements, etc.).
But one major key is the increase in pricing transparency since 1996. In 1996, there were very few outlets available for the average customer to book a hotel room, and today hoteliers must manage different rates across a number of platforms. Because any potential guest can so easily compare rates among a variety of hotels in an instant, logically this would suggest a downward pressure on rates, if even simply for the fact that a heightened competitive environment dictates it.
All in all, it seems reasonable that the more information available to the traveler, the fiercer the competition is for that traveler's dollar. Rates still increase when times are good, but it appears that nowadays they just increase less.