It has long been a challenge for hoteliers to forecast demand, with the industry at the mercy of events such as the SARS outbreak and Sept. 11, 2001.
So researchers at commercial real estate brokerage Cushman & Wakefield Ltd. are now spending more time puzzling over hotel data to see whether, outside of those unlikely scenarios, they can more accurately predict future hotel demand.
One variable that they've been toying around with is office demand. They've been crunching numbers around office supply, absorption rates and office vacancy rates and looking for patterns with hotel statistics. And they've been surprised by what they've found.
Take Toronto. On average, the city has generated 56 hotel room nights for every 1,000 square feet of occupied office space, Cushman found in a new report. But what's interesting is how little deviation there is from that average. Between 2001 and 2012 the number never dipped below 50 or above 58.
"I just didn't know what to expect and what amazes me is the consistency in each of the markets," says Charles Suddaby, who leads the Canadian hospitality group at Cushman. "To me this is an eye opener, because I don't think anybody would have perceived there to have been such a strong correlation."
The deviation around the average number of room nights per 1,000 square feet of occupied office space in Toronto is two. "Consistently … it's 56 to 58 rooms per square foot. Just about every year. I never would have thunk it," Mr. Suddaby says.
"We recognize that individual hotel performance is linked to a broad range of factors such as location, quality, size, brand, level of service and price points that all contribute to its ability to capture a larger share of the market," adds the Cushman report. "However, it is also important to examine the broader market in which a hotel operates in order to gain an understanding of the market's long-term prospects …
"We were able to uncover clear evidence that hotel demand and occupied office space are linked."
While the two variables tend to move together within cities, the average number of hotel room nights per 1,000 square feet of occupied space varies notably between cities, a finding that speaks to the many other elements that are at play in determining hotel demand. For instance, the average number of used hotel rooms per 1,000 square feet of office space in Los Angeles is a whopping 1,092, while in Boston it's 77.
"Interestingly, Vancouver, with many of the same characteristics as Los Angeles, but much smaller in size, received only 144 occupied hotel rooms per 1,000 square foot of occupied office space compared with Los Angeles at over seven times the volume," the report notes.
Calgary's average was 60, Ottawa's 77, and Montreal's 84.
The results of this research raise as many questions as they answer.
"In my naivety, I would have said that Toronto and Boston – with very similar amounts of occupied office space, both strong markets, pseudo gateway markets, strong convention markets, strong leisure destinations – would be fairly similar," Mr. Suddaby says. "So it's quite interesting that there's this difference."
The report notes that while hotel room sales tend to be very short-term and unpredictable, office tenants generally sign long leases.
"In markets where there is substantial office development, we believe that measuring office supply, absorption rates, vacancy rates and other such factors can be of assistance as a means of forecasting hotel demand," it says.
"There are a bunch of factors that influence hotel demand and I think this is the first kick at the can at this sort of analysis," Mr. Suddaby adds. "Trying to project hotel demand is in some ways a bit of a crapshoot. … The commercial sector is only part of the equation, but it is remarkable findings I think."