Airbnb no threat to hotels’ higher yielding compression nights: STR

23 May, 2017

The battle between short term holiday letting communities and traditional hotels has been raging for some time - but new findings reveal that the market might just be big enough for both. Global hospitality benchmarking organisation STR recently highlighted some interesting facts about the accommodation industry, such as the impact of Airbnb on average daily rates, occupancy, as well as supply and demand.

In the data (Airbnb & Hotel Performance: An analysis of proprietary data in 13 global markets), STR analysed information from 13 key cities around the world, with Sydney representing the South Pacific. The 13 markets included: Barcelona; Boston; London; Los Angeles; Mexico City; Miami; New Orleans; Paris; San Francisco; Seattle; Sydney; Tokyo; and Washington, D.C.

Compression Nights under threat - or not?

Since its inception, it has been suggested that Airbnb is negatively impacting hotels on the nights which they make most of their money, i.e. compression nights. A compression night, is where more than 95% of rooms are occupied and in turn, hotels are able to raise their rates, earning more money for the same service.

The implication was that Airbnb reduced the amount of available compression nights, by increasing supply, and offering the same service at a reduced rate. What STR found challenged this conclusion.

Compression Nights for 7 US markets: July YTD

Source: STR

STR focussed its analysis on the seven US cities and identified that the amount of compression nights has been steadily increasing since 2005 when it was only 15 in the year. The number reached its highest in Jul-2015 at 76 nights, reducing slightly in Jul-2016 to 71. Airbnb launched in San Francisco in August of 2008.

What's really interesting is that during this time, the seven markets were seemingly not in fact negatively impacted, with average daily rates on compression nights in 2016 recording a 35% increase on non-compression nights. According to STR this margin has never been achieved before. The six international markets recorded similar results regarding compression night performance.

Occupancy rates reflect hotel occupancy

STR also reviewed occupancy rates across the 13 cities which unsurprisingly had hotels far outperforming Airbnb in the 12 months ending Jul-2016. What STR learnt from this data reaffirms the common belief that solid Airbnb occupancy usually reflects solid hotel occupancy.

Hotel and Airbnb Occupancy: Jul-2016, 12-Month Moving Average

Source: STR

Supply - growing fast

As previously reported by the Blue Swan Daily, in New Zealand, Airbnb has over 20,000 listings, while Sydney alone has passed that number, and is ranked as one of the top 10 Airbnb cities in the world (although this varies seasonally). In 2016, an Airbnb report identified that there were more than 26,000 listings across NSW alone, and some 18,000 of those were in Sydney - doubling from the previous year. A summary compiled by listings.csv.gz showed that in early Apr-2017 there were over 24,000 listings in Sydney - up by 30% again.

These figures do not seem to be impacting average daily rates for each location with the cost of hotels continuing to rise in Auckland despite Airbnb's presence. In fact, Blue Swan's research has clearly shown the average daily room rate for a hotel in Auckland's CBD is close to (and in some cases exceeding) NZD400.

Where the hotel industry is really feeling the effects comes more from the average length of an Airbnb stay, which in Sydney is approximately five nights. For the full 13 markets reviewed, STR found that half of Airbnb bookings were in fact seven days or longer, a massive multiplier.

Other interesting findings for the 12 months ending Jul-2016

Airbnb:

  • 12 out of 13 markets reported supply increases of more than 30%;
  • All markets reported demand growth of more than 20%;
  • Share of market demand was generally below 4%;
  • Share of revenues was below 3%;
  • Half of Airbnb bookings were seven days or longer;
  • Over the period Airbnb average daily rates decreased in eight markets and increased in five; and
  • Share of leisure travel far outweighs business travel (ie the higher value end of the market is less affected).

Hotels:

  • Hotel average daily rates generally were higher than Airbnb rates; and
  • Hotel average daily rates increased in twelve markets and decreased in one.

Inevitably each city has its own characteristics, but the STR findings suggest the impact on hotels tends towards Airbnb generally playing a largely complementary role rather than a diversionary one.

The battle will continue, but the value of studies like this is to provide some statistical background to the debate.