An analysis of total U.S. performance and a peek at several global markets shows that RevPAR changes greater than 10% have happened almost daily since the mid-2000s, which makes volatility for hotel revenue managers part of the normal routine.
HENDERSONVILLE, Tennessee—At STR, we deal in averages. The foundation of the company is to never release individual property data, so therefore our results are always shown as the combination of the performance results of a group of hotels, normally as averages. (STR is the parent company of Hotel News Now.)
Naturally, these averages hide many of the underlying oscillations of the data. It is intuitive that a lot of the daily data points would vary, sometimes wildly, from the average that we report. On a national level, these outliers around the midpoint are pretty small:
The monthly data is, of course, a result of the combination of daily data. It is probably fair to assume then that the daily result fluctuations around the monthly midpoint are probably even more pronounced. We can see this example when looking at the first two months of this year:
As the time unit changes from “month” to “day,” the data gets more volatile. This raised the question: “What is normal?” In other words, the hotel business is a daily—or rather, nightly—business, and decisions about rate and accompanying occupancy are made daily by revenue managers across the land. So, we wondered how common double-digit revenue per available room changes are; in other words, are the wide fluctuations we observed this year outliers or the norm?
A detailed look at daily RevPAR percent changes back to 2005 shows that even for a sample size as wide as the U.S., almost 30% of all days show RevPAR fluctuations of either more than 10% growth or less than 10% decline. The following chart shows the daily RevPAR percent change for the U.S., with the positive outliers in green and negative outliers in red.
The chart looks at the daily RevPAR percent change when comparing day to day, so the first Monday in July compares to the first Monday in July in the prior year. In contrast, the Fourth of July moves days and therefore the performance results vary drastically on the day that 4 July falls. A few takeaways from the national chart:
- The Great Recession of 2008 and 2009 is clearly visible. RevPAR dropped precipitously for that time period, but the recovery period in 2010 did not show very positive RevPAR changes (change was positive, but often not more than +10%).
- There is a wide variety of daily performance throughout the year and seems to be no clear pattern with regards to daily performance variations by season. But when there is a double-digit gain or drop, the reason could be the move of a holiday that is date-specific.
- It is probably fair to assume that the fluctuations are driven by calendar shifts, non-recurring group demand and one-off events.
We then examined the RevPAR fluctuations by day for a variety of markets. The table below shows some of the markets where daily RevPAR changes are above or below 10% for more than half of the days. Results like that certainly do not make revenue management in the specific markets easy. The question of: “What is normal?” is hard to answer for revenue managers and general managers alike in areas where the outlier seems to be the norm.
In other words, if you operating in markets such as San Francisco or Denver, there is a high likelihood that the daily performance of the market, and likely the comp set, varies by more than double digits.
Looking around the world, the wide ranges in RevPAR changes are similar:
Here is the visual display of a few of these markets.
New York City:
The percentage of days that fluctuate in the double digits is the average over a number of years, and since hotel industry performance fluctuates with general economic conditions, it is not surprising that the annual averages also show spikes as the economy deteriorates. The table below breaks out the U.S. performance by year and shows the percentage of days each year that recorded double-digit RevPAR changes.
The roughly 30% number shown above—an average—hides the fact that volatility has decreased somewhat since 2015 and that now between 10% and 20% of days have extreme volatility.
Examining annual data for San Francisco and New York City shows that over the last few years, volatility has remained very high in San Francisco and somewhat ebbed in New York.
The main takeaway from this examination is that wide outliers to the RevPAR average are the norm. It will be helpful for operators in each market to understand if market performance fluctuates more wildly than in other markets and if they can therefore expect more volatile results week to week and day to day. This in turn might be an indicator of how much their daily comp-set data varies from the market results.
Jan Freitag is the SVP of lodging insights at STR. M. Brian Riley is a research analyst with STR’s Market Insights division.
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.