This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies.  Find out more here  Close
Measuring variance in hotel market forecasts
January 23 2013

A look back at STR’s “Market Forecast Reports” shows generally little variance between predictions and actual results in the top 25 U.S. markets.

Highlights
  • The majority of the market projections for supply were high because supply growth unexpectedly was suppressed during the past year.
  • The strength of demand in the industry continues to surprise and contradict many discouraging economic indicators and expert outlooks.
  • San Francisco is the only market where the ADR forecast varied by more than 5%.
By Alex Smith
STR columnist

HENDERSONVILLE, Tennessee—Now that STR, parent company of HotelNewsNow.com, and Tourism Economics have produced the “Market Forecast Reports” for a full year, we wanted to put together a brief summary of how reliable the reports have been for their respective markets.

The first reports were released in August 2011 and contained historical data through the second quarter of 2011. This review covers the first year of projections released in those reports (third quarter 2011 through second quarter 2012).

Although we are very pleased with the  results of this analysis, we firmly believe that over time these reports will become stronger with practice, especially as our knowledge of each market grows.

Forecast methodology
STR partnered with Tourism Economics, an Oxford Economics company with a singular focus on quantitative analysis of the travel industry, to build econometric models unique to each market for these reports. Robust equations have been econometrically estimated that closely follow past movements in hotel performance as measured by STR data. These equations are used to forecast hotel performance using economic forecasts from Oxford Economics' global macroeconomic database as well as Oxford Economics' global city and region forecasts. Economic forecasts are augmented with specific intelligence to determine the additional effect on hotel demand and ADR of any events hosted within that market.

Analysis
In this review, we have calculated the percentage variance for the aggregated first four quarters that were forecasted in August 2011 (third quarter 2011, fourth quarter 2011, first quarter 2012 and second quarter 2012). The variance is calculated as follows:

Percentage Variance = (Forecasted Value – Actual Value)/Actual Value*100

Therefore, positive values indicate the forecast over-predicted the performance for that metric, and a negative value indicates under-prediction.

Results
The majority of our market projections for supply were high because supply growth unexpectedly was suppressed during the past year.

This factor combined with the under-predictions of demand in 15 markets (with Nashville, Tennessee, and Houston being the most significant) led to our occupancy forecasts being low in all but five markets. The strength of demand in the industry continues to surprise and contradict many discouraging economic indicators and expert outlooks. When these forecasts were released last year, a much greater slowdown in demand was anticipated for these markets.  

Given that average daily rates are dependent on other performance factors and are very difficult to predict, we are pleased with the general accuracy of our ADR forecasts. San Francisco is the only market where the ADR forecast varied by more than 5%. The fantastic pace of rate growth that has occurred in this market was unanticipated. The markets are roughly split between being over- and under-predicted, and all have relatively low variance levels.  

Finally, taking the forecasted metrics to the bottom line and looking at revenue per available room, the top 25 markets generally have outperformed expectations just as the total industry has. Our RevPAR projections for 16 markets came out below the actual results, and all of the most significant errors were in markets that have greatly overachieved. These markets include: Nashville; Oahu Island, Hawaii; Tampa, Florida; Houston; New York; and San Francisco. With the exception of San Francisco, these errors all stemmed from under-predicted occupancy, mostly the consequence of low demand forecasts.

Click image to enlarge.


 

COMMENTS   Show All
Login or enter a name   Post Your Comment  Check to follow this thread via email alerts (must be logged in)
(4000 characters max)

Comments that include links or URLs will be removed to avoid instances of spam. Also, comments that include profanity, lewdness, personal attacks, solicitations or advertising, or other similarly inappropriate or offensive comments or material will be removed from the site. You are fully responsible for the content you post. The opinions expressed in comments do not necessarily reflect the opinions of Hotel News Now or its parent company, STR and its affiliated companies. Please report any violations to our editorial staff

TRENDING
Airbnb accounts for 5.4% of NYC demand
AC brand enters US, 50 more on tap
NYC Plaza saga continues to unwind
IHG’s Kimpton buy a ‘perfect fit,’ CEO says
Freitag’s forecast: 10 predictions for 2015
6 things about Expedia you didn’t know
VIDEO
Wyndham focused on management
Greg Mount talks RLHC's turnaround
The Lodging Conference in review: Day Two
The Lodging Conference in review: Day One
US supply growth not a problem
LATEST NEWS
STR: US hotel performance for November 2014
Destination Hotels puts on new face for guests
AH&LA, AAHOA throw down gauntlet on City of LA
How to prevent, handle employee theft
GM finds closure in New Orleans
Freitag’s forecast: 10 predictions for 2015
Contact Us
Hotel News Now
18500 Lake Rd.
Suite 310
Rocky River, Ohio 44116
        
Copyright © 2004 - 2014 Hotel News Now, a division of STR, Inc. All Rights Reserved.   Privacy