International Journal of Hospitality Management 73 (2018) 85–92

Contents lists available at ScienceDirect

International Journal of Hospitality Management

journal homepage: www.elsevier.com/locate/ijhm

Airbnb’s effect on hotel sales growth T

a b,⁎ c
Inès Blal , Manisha Singal , Jonathan Templin
a
Ecole hoteliere de Lausanne, HES-SO // University of Applied Sciences Western Switzerland, CP 37- Route de Cojonnex 18, 1000 Lausanne 25, Switzerland
b
Department of Hospitality and Tourism Management, Virginia Tech, Pamplin College of Business, 362 Wallace Hall, 295 W. Campus Drive, Blacksburg, VA 24060, USA
c
Department of Educational Psychology and Achievement and Assessment Institute, University of Kansas, 614 Joseph R. Pearson Hall, 1122 West Campus Road, Lawrence,
Kansas 66049, USA

A R T I C L E I N F O A B S T R A C T

Keywords: Disruption of existing industries by peer-to-peer sharing economy platforms, like Airbnb, has received extensive
Sharing economy media attention primarily regarding the regulatory and the social aspects of the sharing economy. In this paper,
Airbnb we examine the substitution and complementary effects of Airbnb supply on hotel sales performance patterns in
Lodging San Francisco, the birthplace of Airbnb. Our results, based on a mixed-model analysis using a saturated, un-
Sales performance
structured covariance matrix, show that overall hotel RevPAR is unrelated to total Airbnb supply. Interestingly
Disruptive innovation
however, in certain segments, RevPAR is affected by the average price of Airbnb listings. More importantly, hotel
sales performance is impacted by Airbnb customer reviews, which points to nuanced and contextual com-
plementary and substitution effects. We outline suggestions for future research as well as practical implications
for hotel firms operating in similar markets where Airbnb has high penetration rates.

1. Introduction consumer demand angle trying to profile users (Guttentag et al., 2018),
their motivations (Tussyadiah, 2015), and their satisfaction with Airbnb
Epitomized by Airbnb, the sharing economy, which is driven by versus traditional hotels (Guttentag and Smith, 2017). Yet, research
two-sided platform information technology, and a desire to monetize that examines the direct impact of Airbnb supply on incumbent hotels is
underutilized personal assets such as a house, has taken a firm hold on scant (Choi et al., 2015; Dogru et al., 2017; Gutierrez et al., 2017;
consumer habits. While homes have been used for short-term vacation Neeser, 2015; Schneiderman, 2014; Zervas et al., 2017). Most studies
rentals for quite some time, Airbnb allows hosts and guests to list and rely on cross-sectional data (Zervas et al., 2017; Zervas et al., 2017 and
book accommodations around the world via an online community Neeser, 2015 are exceptions), and thus do not infer the effect of
marketplace (please see www.Airbnb.com). In 2018, a short ten years Airbnb’s trajectory of growth and the satisfaction of its users on hotel
since its creation, Airbnb is valued at $31 billion, is present in over performance metrics over time. Furthermore, Guttentag and Smith
65,000 cities in 191 countries and boasts over 3 million users. The (2017), in their study on demand side perception of substitute effects of
advent of Airbnb in 2008 and its rapid growth has captured, not only, hotel versus Airbnb users, also urge future research to use longitudinal
media attention but has been the subject of several industry reports like data to understand if and when Airbnb is used as a substitute for certain
the American Hotel and Lodging Association (AHLA) Report (O’Neill types of accommodation. To answer the call for research in this area, we
and Ouyang, 2016) and CBRE Hotels Research Report (Lane and attempt to fill the gap by examining how Airbnb supply, price, and
Woodworth, 2016). As such, it has also been a topic of hot debate at perceived quality affect incumbent hotel sales performance long-
several hotel conferences, such as the Hotel Investment Conference, itudinally in the San Francisco market. We choose the San Francisco
Hotel Sales and Marketing Association International, Travel and market because it is where Airbnb was founded and is one of its largest
Tourism Research Association over the last few years, often with topics − and more importantly, oldest − markets.
such as “disrupting the disruptor”. Despite the company’s claim that it is a mere supplement to hotels’
Academic research has primarily studied Airbnb’s regulatory or services (Gallagher, 2017), our overarching proposition is that Airbnb
public policy issues (Kaplan and Nadler, 2015) as well as the social offer has an impact on incumbents and that this effect is more complex
aspects of sharing (Edelman et al., 2017; Kakar et al., 2016; Kim et al., that supplementing the supply of rooms. Therefore, based on the con-
2015). In recent years, research has looked at the micro angle: i.e., host ceptual framework of disruptive innovation (Bower and Christensen
motivations for listing on Airbnb (Karlsson and Dolnicar, 2016), or the 1995; Guttentag, 2015), and using theoretical arguments on new

⁎
Corresponding author.
E-mail addresses: [email protected] (I. Blal), [email protected] (M. Singal), [email protected] (J. Templin).

https://doi.org/10.1016/j.ijhm.2018.02.006
Received 25 July 2017; Received in revised form 4 February 2018; Accepted 6 February 2018
0278-4319/ Published by Elsevier Ltd.
I. Blal et al. International Journal of Hospitality Management 73 (2018) 85–92

entrants into an industry and their effects on incumbents from the although it could possess some attributes that straddle technological or
strategic management literature (Geroski, 1995; Seamans and Zhu, radical innovation.
2014), we explore whether 1) the development of Airbnb supply and A business model innovation redefines an existing product/service
users’ satisfaction have an effect on hotel sales patterns, and 2) whether and the manner in which it is provided to the customer. A new business
there are differential effects of the impact of Airbnb based on hotel model invades a current market by highlighting different qualities and
segments. features offered by current firms, thus attracting a different customer
Our analyses, based on 1111 observations, find that while total base and enlarging the existing economic pie. Because the customers
Airbnb listings (or total inventory) do not impact the growth trajectory are different, and the supply chain activities of the business model in-
of hotel RevPAR (Revenue Per Available Room, the standard measure of novation are incongruent with the traditional value chain, incumbent
hotel sales performance), thus confirming the supplementary service firms do not modify their activities or respond to the new entrant
argument, the average price of an Airbnb offer is positively associated (Markides, 2006). This has indeed been the stance taken by Airbnb all
with patterns of hotel sales performance. In line with the general along. The company maintained that not only did they cater to the
finding from Zervas et al. (2017), we find that the effects of Airbnb vary price-conscious customer but to a niche market of even those customers
across different segments of the industry, however in a manner different who wanted the “local, home, immersive experience”, and who would
from the Texas market. Our most interesting result is that the sa- not otherwise travel to stay in traditional hotels (Lawler, 2012;
tisfaction of Airbnb users is negatively associated with hotel perfor- Shankland, 2013; Yglesias, 2012). Incumbent firms initially dismissed
mance patterns. In other words, our results show that the increase in the Airbnb as a fringe player unable to foresee its later appeal to a main-
quality of Airbnb service directly impacts hotel performance adversely, stream consumer base. According to Markides (2006), as time passes,
thus indicating substitution effects. the business model innovation improves to enable parity with existing
The rest of the paper is structured as follows: first, we discuss the features as well as superiority over new features thus attracting more
theoretical background based on past literature in order to develop our customers both old and new. In addition, the innovation then provides
proposed hypotheses. In particular, we present current knowledge on more than just a supplement to the existing offer. As a result, it attracts
disruptors’ impact on incumbents with a focus on the hotel industry. the attention of the media and established players. At this stage, in-
The following section outlines our methodology, describing the sample cumbent firms may attempt to thwart disruption by preventing the
used and the saturated unstructured covariance model employed to innovator from entering their market. To do so, they might leverage
analyze our repeated measures of hotel sales performance. We then their well-established networks or try to persuade public authorities to
report and discuss the results of our analyses, and draw conclusions, enact regulation. Indeed, in several states like New York and California,
explaining the implications for practice, the limitations of the study and laws have been passed regulating Airbnb, oftentimes at the behest of
several suggestions for future inquiry. hotel industry associations (Benner, 2017). The American Hotel and
Lodging Association (AHLA) has lobbied at several fora to prevent,
2. Theoretical background and hypotheses development regulate, and tax Airbnb operators to level the playing field with their
members.
2.1. Airbnb as disruptive innovation Using the ultimate asset-light business model - where Airbnb owns
no real estate and charges only fees to its users - it has scaled up supply
San Francisco was the first city in which Airbnb evaluated its eco- in an industry where barriers to entry could be considered high because
nomic effect. This impact study, conducted in 2012, found that 72 of, notably, high capital investment requirements. As it stands, the
percent of Airbnb properties in San Francisco were located outside the supply growth in the industry continues to increase and demand growth
central hotel district and Airbnb generated approximately $56 million has been partially able to mitigate the impact of new supply growth of
in local spending. It also reported that while the average San Francisco hotel rooms. Recently however, occupancy rates in large markets in-
hotel guest visited for 3.5 days and spent $840 per visit, the average San cluding San Francisco have begun to decline. Rael (2017), in a study
Francisco Airbnb guest visited for 5.5 days and spent $1045 (Geron, conducted on the United States’ top 25 markets, reports that when the
2012) suggesting a supplementary role of the new offer. The impact on supply of hotel rooms grows at a rate of at least 1.8%, occupancy begins
incumbent hotels was either not evaluated or not reported in this study to slip; and when supply expands at a rate of at least 3.4%, RevPAR
of the San Francisco market. begins to decline. While these numbers pertain to the supply of hotel
Airbnb, along with other sharing economy firms, has been described rooms, the supply of Airbnb rooms (as clear – albeit imperfect – sub-
as mostly − but not completely − falling within the definition of stitutes) would also seemingly dampen hotel occupancy and sales per-
disruptive innovation (Guttentag and Smith, 2017). In their seminal formance indicators. This is especially true given that Airbnb continues
paper on disruption, Bower and Christensen (1995) describe a dis- to refine its features in an effort to offset its limitations when compared
ruptive innovation as occurring outside the value network of existing to the traditional hotel product.
firms, and one that introduces a different package of attributes offered When Airbnb began operations there were stories of vandalism and
to and valued by customers that eventually transforms a product or issues of safety, which the company resolved with blanket insurance to
service, to the extent of overthrowing incumbent dominant firms in the hosts, thus ensuring that supply constraints would not affect its own
industry. The product/service benefits may focus on providing con- growth. To alleviate safety and credibility issues, Airbnb provides ser-
sumers with a simpler, more convenient, and often a more price-effi- vices like photographs that enhance its authenticity (Ert et al., 2016),
cient experience. Christensen (1997) describes the process of disruptive and a mutual rating system of host and customer – a key attribute of the
innovation as occurring in phases: an initial step when the innovation is sharing economy. Although some scholars believe that it is overly op-
not adopted by mainstream customers, performs worse than existing timistic (Zervas et al., 2015), the rating system still serves as a proxy for
products, and is used only in niche or fringe markets; in a later stage, reputation of both the host and the guest. The exponential growth and
mainstream customers adopt the innovation rapidly, thus disrupting the acceptance of Airbnb, which now has more room inventory than the
leading firms in the market. Christensen (1997) studied disruption in largest hotel players combined (Hartmans, 2017) and offer a variety of
the context of miniature hard drives, and since then, several industries accommodations from a shared room to a private island and everything
have seen disruptive innovation not only in products like DVD/Blu-ray, in between, has transformed it into a global player, with a presence in
but also in services like taxis (Uber), or classified ads (Craig’s list), and over 190 countries and a user base of over 4 million as of November
short-term accommodation with Airbnb. Although scholars debate be- 2017 (airbnb.com). In any case, it has gone well beyond the first stage
tween types of disruptive innovation (Danneels, 2004; Markides, 2006), of disruptive innovation. Therefore, our focus in this paper is to explore
we discuss Airbnb from the angle of business model innovation, whether Airbnb supply presents the potential to substitute for the

86
I. Blal et al. International Journal of Hospitality Management 73 (2018) 85–92

leading hotel firms especially as its hosts begin to mimic hotels offering guests total 160 million worldwide) a ‘supplementary service’ or ‘en-
greater amenities and concierge-like services (www.airbnb.com). larging the pie’ argument would suggest that Airbnb has a marginal
While reviewing data from Airbnb listings in 12 major metropolitan effect on hotel prices. This is because these customers are different and
U.S. cities from September 2014 through September 2015, O’Neill and unlikely to also be hotel customers. In the Nordic countries of Norway,
Ouyang (2016) in a study funded by the AHLA, found that contrary to Finland and Sweden for example, Neeser (2015) finds that Airbnb did
common perceptions (i.e., that Airbnb hosts are homeowners who rent not significantly affect hotels’ average revenue per available room but it
out their homes or part thereof for additional income while they are did contribute to a reduction in the average price of a room in areas
away), 28.5% of the revenues generated were attributed to full-time where Airbnb had higher penetration rates. In their study, the authors
operators (hosts who rent out units for 360 days per year) while multi- observed that the “cultural experience” provided by Airbnb hosts made
unit operators (hosts operating two or more units) generated 39% of Airbnb a better choice for foreigners than locals.
Airbnb revenues. In another study conducted by the New York Attorney However, Guttentag and Smith (2017) report that in their survey of
General’s office in New York City from January 2010 until June 2014, over 800 Airbnb guests, nearly two-thirds had used Airbnb as a hotel
Schneiderman (2014) reported that, there was explosive growth in the substitute. A substitution effect would then suggest that higher Airbnb
short-term rentals market, and that a majority of these rentals violated rates would also result in higher average daily rates for hotels especially
the law (that hosts can rent for 30 days only and must reside on the in large city-markets like San Francisco where Airbnb has a higher
premises while sharing their accommodation). The study also found penetration rate than smaller markets. We thus propose that:
that while multi-unit owners accounted for only 6% of total hosts in the
H2. Average prices of Airbnb rentals are positively associated with
study, these hosts ran multi-million dollar businesses, a dispropor-
hotel sales pattern performance
tionate share by volume and revenue. This new structure of the Airbnb
supply could suggest that the company’s offer may be more than a As noted above, the features offered by Airbnb differ from the tra-
supplement to hotels in a destination. ditional hotel product. Studies that profile Airbnb customers report that
Similarly, Lane and Woodworth (2016) examined 59 cities encom- they are more likely to be early adopters, extraverted, interested in
passing 229 submarkets in the U.S. between October 2014 to September local culture and sights. They are also likely to spend more time plan-
2015 reporting that the proportion of revenue generated by multi-unit ning holidays than people who stay in hotels (Hajibaba and Dolnicar,
operators who have business-like operations is very large and they are 2017). Yet two-thirds of Airbnb users rented an Airbnb unit as a hotel
more likely to be in competition with the hotel sector. In fact, New substitute, not only at the lower end but also substituting a mid-range
York, Los Angeles, San Francisco, Miami and Boston (the five biggest or upscale hotel with an entire home rental on Airbnb (Guttentag and
cities in the U.S.) accounted for more than 55 percent of Airbnb’s $2.4 Smith, 2017). This trend then indicates that if the quality of Airbnb
billion in revenues. The authors developed an Airbnb Competition accommodation and user’s satisfaction with their Airbnb experience is
Index, a measure of risk that incorporates comparisons of Airbnb’s met or exceeded, hotel sales performance would be adversely impacted.
Average Daily Room rates (ADR) with those of traditional hotels; the Part of the ease of substitution results from low switching and
scale of the active Airbnb inventory in a market to the supply of tra- transaction costs. Opening an account on Airbnb and searching for
ditional hotels, and the overall growth of active Airbnb supply in that accommodation is relatively easy and mimics searches on Online Travel
market. San Francisco was second, at 81.4, only to New York regarding Agent (OTA) websites, which are typically used by customers when
the risk index from growth of Airbnb. In 2015, according to this report, booking hotels. While Airbnb transaction costs could be high due to
the Q4 hotel occupancy rate in San Francisco was 84.5%, while the lack of transparency and information asymmetry, identification and the
apartment unit supply stood at 216,337, with a vacancy rate of 3.5% dual rating system − where both hosts and guests rate each other after
(Lane and Woodworth, 2016) indicating an impact of Airbnb supply on every transaction − mitigates some of the trust limitations (Allen and
hotel performance. Berg, 2014).
In a more comprehensive study of the Texas market, Zervas et al. To attract segments that do not fall into the category of the typical
(2017), found that in the Austin market, where Airbnb has the highest Airbnb leisure traveler, such as business travelers, Airbnb has actively
supply/penetration rate, hotel revenue is negatively impacted by collaborated with convention bureaus, conference hosts, and state
8–10%. We believe that in the San Francisco market, the effect will be governments to expand their reach (Hospitality Biz India, 2017). The
similar, and hence hypothesize that: company has also actively sought to provide accommodation in times of
crises such as after natural disasters including Hurricane Sandy and the
H1. Total Airbnb supply is negatively associated with hotel sales
Seattle earthquake (Nickelsberg, 2014). Consumer perceptions of
pattern performance (RevPAR)
Airbnb as a socially-oriented company may also play a role in higher
satisfaction ratings. We thus hypothesize that:
2.2. Substitution effects based on price and quality
H3. Average satisfaction of Airbnb users is negatively associated with
hotels sales pattern performance
Although Airbnb itself claims that there are likely no substitution
effects between its services and hotels (Gallagher, 2017), empirical
research has provided mixed results. With increased supply, as pro- 2.3. Differential effects based on heterogeneity of hotel segments
posed in H1, there is likely to be downward pressure on the average
price of hotels. Zervas et al. (2017), report that with the entry of Airbnb Based on the logic of the hypotheses derived above, a new market
not only are hotel revenues affected but this impact manifests itself entrant will generally cause a negative outcome for incumbent firms in
primarily through less aggressive hotel pricing (i.e. even during periods the industry and create at least some substitution effects, despite new or
of peak demand hotels’ ability to raise prices is tempered by Airbnb additional features of the product or service offered (Geroski, 1995;
supply, thus reducing hotels’ value capture). Seamans and Zhu, 2014). However, not all incumbent firms will be
At the same time, Lane and Woodworth (2016) found a correlation equally affected as the disruptive effects of the new entrant depend on
between the higher prices of hotel rooms and the number of Airbnb the context in which the entry occurs, and on the particular attributes of
units available, partially feeding into the “enlarging the pie” argument the incumbent firms. Past literature has shown that the effect of addi-
(Pizam, 2014). This also illustrates that Airbnb hosts respond to market tional supply or competition provided by an entrant will vary across
incentives and that the effects of the sharing economy on the hotel hotel segments. In fact, Kosová and Enz (2012) find differential effects
business might be more intricate than expected. As acceptance and of crises on room sales based on the size of hotel, while O’Neill and Xiao
adoption of Airbnb continues (listings currently stand at 3 million and (2006) observe different effects of brand affiliation on value, based on

87
I. Blal et al. International Journal of Hospitality Management 73 (2018) 85–92

hotel segments. Blal and Sturman (2014), find that the quality and Table 1
quantity of online reviews have different impacts on hotel sales in Sample size per segment.
London based on whether the hotel is lower or higher tier. Thus, as
Segment Number of hotels per segment
previous research has shown (Zervas et al., 2017), not all hotel seg-
ments will be equally affected by Airbnb (i.e., there will be differential Luxury 12
effects on sales performance based on the heterogeneity of hotel seg- Upper Upscale 8
Upscale 14
ments).
Midscale 41
The threat of entry and growth of Airbnb is directly felt by low-end Economy 26
hotels and traditional B&Bs because private room prices are generally TOTAL 101
on par with rooms offered by 1- and 2-star hotels (Guttentag, 2015).
Hajibaba and Dolnicar (2017) surveyed 689 travelers in Australia to
measure their perception regarding peer-to-peer accommodation’s Table 2
Description of San Francisco hotel data.
likelihood to replace different segments of already established com-
mercial options. Their results show that low-end accommodation pro- HOTELS
viders such as 1-star and unstarred hotels were most in danger of being
replaced by peer-to-peer accommodation network offers while 4- and 5- Date Total Room Average ADR Average Average
star hotels were the least likely to be affected…at least for now. Supply ($) Occupancy rate RevPAR ($)

Similarly, in Texas it was found that a 1% increase in Airbnb’s 07.Dec.2013 39′092 171.81 93.58 161.04
listings resulted in a 0.05% decrease in quarterly hotel revenues (Zervas 12.May.2014 39′229 202.17 88.69 179.10
et al., 2017). Although this effect is less than a comparable effect of a 25.Aug.2014 39′205 312.47 95.84 299.42
0.29% decrease in revenue if the supply of hotel rooms increased, the 19.Feb.2015 39′030 181.30 78.08 141.88
06.Jun.2015 39′047 209.17 86.44 182.59
exponential growth of Airbnb listings renders the small decrease sub-
21.Aug.2015 39′227 238.52 90.16 215.01
stantive (Zervas et al., 2017). The study by Zervas et al. (2017) also 21.Oct.2015 39′314 268.46 94.56 254.61
found that while Airbnb penetration is negatively correlated with hotel 22.Nov.2015 39′150 153.33 55.60 85.22
revenue, this effect is different across hotel subsectors. Hotels catering 14.Dec.2015 39′150 219.26 88.62 194.87
to business travelers and upper-scale hotels were less affected. 17.Jan.2016 39′361 190.96 83.56 159.53
18.Feb.2016 38′084 204.95 80.80 165.82
We therefore propose that
H4. The effects of Airbnb on hotel sales pattern performance varies
across different hotel segments Table 3
Descriptive of San Francisco Airbnb data.

3. Methodology AIRBNB

Date Total Total Average Average Average

To explore the effects of Airbnb’s growth trajectory on hotel sales listings reviews number of satisfaction Price
performance, we estimate a series of mixed models on data collected reviews score
over 11 occurrences from 101 hotels. In these models, occasions are
07.Dec.2013 110 965 9 4.75 244.64
nested within hotels and hotels are nested within segments or neigh-
12.May.2014 4′275 31′469 12 4.79 204.60
borhoods. In this repeated measure mixed model analysis, we focus on 25.Aug.2014 3′312 26′088 14 4.79 209.26
the trend of change in hotel sales performance over time as Airbnb 19.Feb.2015 5′202 34′672 10 4.61 253.99
develops its supply in San Francisco. Our approach, thus complements 06.Jun.2015 5′305 62′408 19 4.76 508.00
21.Aug.2015 5′140 59′870 20 4.76 631.07
past studies that rely either on cross-sectional data or a before entry/
21.Oct.2015 3′211 35′797 19 4.77 288.94
after entry design. Furthermore, we include three dimensions in our 22.Nov.2015 3′182 35′693 19 4.76 276.82
exploration of the effects of this innovation on the hotel industry. We 14.Dec.2015 7′396 80′413 19 4.76 293.19
not only investigate the effects of the total inventory (i.e. number of 17.Jan.2016 6′026 65′460 19 4.76 320.44
listings) of Airbnb offers on the evolution of hotel performance, but we 18.Feb.2016 6′104 104′139 17 4.74 266.56

also explore the effects of the price of Airbnb properties, and more
importantly the effect of Airbnb users’ satisfaction (i.e. Airbnb guest’s
reviews per property listed, the average satisfaction score given by
reviews of the listed property) on hotel sales patterns to estimate the
guest to the host regarding his/her property, and the average rate of
nature of substitution that Airbnb provides in the lodging market.
these properties).
3.1. Sample and data
3.2. Model analyses
We obtained anonymized data from Smith Travel Research (STR)
for 101 hotels in San Francisco for the period between December 2013 While past studies on Airbnb’s effects on hotel performance relied
and February 2016. The repeated measures mixed model of our study is on cross-sectional data and analyses, with a before-after design, we
based on 11 occurrences and the hotels in the sample ranged from the focus here on the trend of change of hotel sales over time and tested
economy to the luxury segment (Table 1). To maintain the con- how Airbnb might affect it. In other words, in this paper, we focus on
fidentiality of the hotels in the sample, property data was coded and how hotel sales performance changed over time as Airbnb entered a
new identification numbers were generated. specific market and as it expanded its business there. Moreover, we not
We used the hotels’ total revenues, total supply, number of rooms, only examine the supply dimension of Airbnb and its effect on hotel
years since opening and renovation as well as the brand affiliation in- sales performance, but we also analyze the impact of its rate and cus-
formation supplied by STR. The average ADR (Average Daily Rate) over tomer satisfaction (a proxy for perceived benefits) on the pattern of
the examined period was $214 for an occupancy rate of 85% (Table 2). RevPAR.
The data from Airbnb (Table 3) was scraped from their website on We therefore used a mixed model with a saturated unstructured
the 11 occurrences of the sample (i.e. the total number of properties covariance matrix for the relationship of residuals across time to ana-
listed on the platform at a given occurrence, the average number of lyze the growth of RevPAR at 101 individual hotels. The main

88
I. Blal et al. International Journal of Hospitality Management 73 (2018) 85–92

advantage for the unstructured covariance model is that no constraints unemployment rate at the time of measurement of hotel performance
are imposed on the values. In other words, we made no assumption and two dummy variables to control for (ii) the seasonality (1: high
about the patterns of variance and covariance across the different hotels season, 0: low season) and (iii) week-end/week-day effects on the
of the sample to then examine the effects of our dependent variables on RevPAR (1: weekend, 0: weekday), which can affect the RevPAR of
the patterns. It is therefore well-suited when there is no easily-dis- hotels. Additionally, we used total hotel room supply (Av_Hotelsupply)
cernable pattern in the data and when data fluctuates over time such as in the neighborhood. The variable had a significant, and trending, main
hotel RevPar (Hoffman, 2015). We applied the model to our sample of effect on RevPAR.
11 repeated measures of individual hotel RevPar between December In line with previous research in hospitality (Tsai et al., 2006;
2013 and February 2016. Zervas et al., 2017), a significant main effect of unemployment was
As the data were collected over 11 occasions from 101 hotels, a found (β = −10.97, p < 0.001) indicating that RevPAR decreased by
series of mixed models were estimated where occasions were nested 10.97 for every additional one percent increase in unemployment. Also,
within hotels and hotels were nested within segments or neighbor- there was a significant main effect of season, where the high season’s
hoods. This repeated measures mixed model provides an integrated mean RevPAR was $45.37 higher than it was during the low season.
approach when studying the structure and indicators of change. In our Additionally, a main effect of ‘day of the week’ was found where
case, we were exploring the structure and indictors of change of in- weekends had a mean RevPAR that was $27.83 higher than on week-
dividual hotel RevPar over time (Hoffman, 2015). As for the saturated days.
model, it includes the main effects of time (i.e. within-subject variance, Furthermore, for the within-hotel effects, we controlled for whether
namely within-hotel variance) and all other main effects, in addition to the hotel in the sample was affiliated with a brand (labeled
the other higher interactions. At higher levels, we tested whether hotel “Operations”). The sample counted 35 chain owned and/or managed
segment, Airbnb supply, Airbnb users’ satisfaction and Airbnb price properties (coded 1), 26 franchised hotels (coded 2), and 40 in-
affected the growth trajectory of hotel sales. dependently-managed units (coded 3). The variable had a non-sig-
In particular, we tested the following model: nificant effect on changes in RevPAR. The two other control variables
also had a non-significant effect on RevPAR. In particular, we con-
Performance THS = π0HS + π1HS*TimeTHS + Hotel LocationHS +
trolled for two aspects at this level. The first, Experience, accounts for
Segment + Airbntotaloffer + AvSat + AvPrice + Control variables + eTHS
the number of years since opening or conversion. Existing research
(McCann and Vroom, 2010) suggests that the experience of the in-
Where: cumbents (i.e., hotels) influences how it would react to new entrants
Performance THS: is the sales performance (i.e. RevPAR) in time T of (i.e., Airbnb). The third variable takes into account the effects of the
a hotel H in the neighborhood where Airbnb is present. The first oc- hotel size (i.e. number of rooms).
currence was the RevPAR on December 7, 2013 (coded 0) and the last Prior to adding our predictors, we first worked on determining the
was the RevPAR on February 18, 2016 (coded 10). number of levels in the model and then sought to determine the trend
Hotel Location: is the neighborhood where the hotel is located across time, if any. To determine the number of levels present in the
(Table 4). data, a two-level empty-means/random intercept model where occa-
Segment corresponds to the category of the hotel in the sample. sions were nested within hotels revealed an intraclass correlation of
Table 1 presents the coding details and the sample size per category. 0.58, which was significantly different from zero (p < 0.0001). Adding
We measured the effects of the platform business with the total neighborhood at level three, however, resulted in a non-significant
number of Airbnb listings available on the company’s website at the random intercept variance of neighborhood. As such, we attempted to
times of measurement. This inventory level approximated the volume use segment as the level three variable as hotels were nested within
effect of the substitute product (coded “Airbnbtotaloffer” and its de- segments. Similarly, this model resulted in a non-significant random
scriptive is in Table 3). The second variable to assess the effect of intercept variance of segment. One factor mitigating the ability to es-
Airbnb, which we used to test for the third hypothesis (H2), is the timate random effects was the sample size at level three, where 10
average price of Airbnb properties listed on the platform at the time of neighborhoods and five segments existed. As such, we decided to re-
the occurrence (“AvPrice”). The third operationalization of the effect of move the level-three random intercept from the modeling process only
Airbnb on hotel sales performance is guest satisfaction (coded “AvSat”). to add the segments variable as a fixed effect when we added our
It is the average of Airbnb guests’ satisfaction with the accommodation predictor variables.
they rented (H3). It is the five-point scale score, which Airbnb guests Next, we sought to determine the trend in RevPAR across occasions,
note when evaluating their stay with the host. We used this variable to if any. Fixed and random linear and polynomial two-level multilevel
operationalize the quality of the substitute. models were used; however, none showed any significantly better fit
The control variables for the within-occasion effect are: (i) the than the saturated, unstructured covariance matrix of RevPAR across all
11 occasions. As such, our analysis focused on fluctuation across time
rather than a time trend (Hoffman, 2015). We proceeded with the un-
Table 4 structured covariance model, a multivariate mixed model, when adding
Neighborhoods in San Francisco: coding and sample size. our predictors to the model.
All quantitative predictors were mean-centered to aid in inter-
Neighborhood Code N*
pretation. The main effects of all predictors were added to the multi-
Castro/Upper Market/Noe Valley 1 1 variate mixed model simultaneously along with interactions between
Chinatown 2 14 our variables of interest and segments. The Kenward-Roger degrees of
Downtown/Civic Center 3 25 freedom method was used to calculate the denominator degrees of
Financial District 4 5
Marina 5 8
freedom for each effect.
North Beach 6 11
Pacific Heights/Western Addition/Presidio 7 5 4. Results
Parkside 8 1
Russian Hill 9 11
The effect of Airbnb in terms of volume (measured as the total
South of Market 10 20
TOTAL 101 number of listings on the platform) predicted in H1, had a non-sig-
nificant effect on RevPAR. In other words, we fail to accept the first
* Number of hotels per neighborhood. hypothesis of an effect of the sheer volume of properties listed on the

89
I. Blal et al. International Journal of Hospitality Management 73 (2018) 85–92

Table 5 data is a reliable source of data and the protocol we used to assess its
Type 3 tests of fixed effects. quality is effective, this method could introduce some discrepancies.
However, it was the only source of data for our study since data from
Effect Numerator DF Denominator DF F Value P-value
Airdna did not cover the early stages of Airbnb in San Francisco.
Segment 4 84 9.15 < 0.001*** Moreover, we used only Airbnb data to examine the effects of the de-
Unemployment 1 99 14.81 < 0.001*** velopment of the sharing economy in the hotel industry, and while it is
Season 1 99 400.96 < 0.001***
the market leader among peer-to-peer accommodation sharing plat-
Weekend/day 1 99 106.41 < 0.001***
Operation 2 84 0.31 0.738 forms, other players (albeit small ones) are also active in the sector.
Size 4 84 0.58 0.681 These platforms may also impact incumbent hotel performance. Future
Experience 1 84 0.31 0.581 research may take a more comprehensive look at all peer-to-peer
Airbnb Total Offer 1 84 0.00 0.956 sharing accommodations to gauge an overall effect of substitution.
AvPrice 1 84 7.25 0.009***
Finally, our findings about the impact of disruptive innovation are
AvSat 1 84 5.69 0.019**
Av Hotel Supply 1 84 4.74 0.032** limited to the specificities of the hotel sector and does not consider
AvPrice*Segment 4 80 3.02 0.023** other commercial accommodations like hostels and Bed and Breakfasts.
In addition, future research may specifically examine issues of compe-
** and *** indicate significance at 5% and 1% levels, respectively. titor analyses and dynamics of threat and response action with regard to
customer co-creation and engagement as a goal that may make hotels
platform on hotel sales in the San Francisco area. These results are in more attractive than their substitutes, thereby increasing switching
line with a recent report on the Airbnb effect in Boston where − despite costs.
tremendous growth in supply of Airbnb due to its flexibility, and a high
occupancy rate increase − hotel prices and revenues did not seem to be 5.1. Contribution and conclusion
adversely impacted (Dogru et al., 2017). Similarly, according to a STR
study of 13 global markets, Airbnb listings did not affect hotel revenues The contribution of this paper is threefold. First it examines the
(Haywood et al., 2017). effects of the Airbnb phenomenon on hotel sales from a longitudinal
However, significant effects were observed for average Airbnb standpoint to test for the impact on Airbnb on the patterns of hotel sales
property price (H2), average Airbnb guest satisfaction (H3), and hotel performance over time. Second, it offers insights for hotel managers
segments as expected in hypothesis 4 (Table 5). Results show that first, facing the development of this new offer in a highly competitive in-
Airbnb property price had a positive effect on hotel RevPar: the higher dustry. Finally, our study contributes to the growing research on the
the price of the rentals posted on the platform, the higher the RevPar of effects of disruptive innovators on incumbents.
hotels. Second, the model analysis confirmed a negative relationship First, the methodology we used focuses on the change in sales
between hotel RevPAR and the average satisfaction rate of Airbnb performance over time rather than a cross-sectional analysis. We used a
guests. Indeed, AvSat had a significant main effect showing a decrease saturated, unstructured covariance matrix to observe the change in
of −25.54 in RevPAR for every unit increase in AvSat. In other words, individual hotels’ RevPAR and how they reacted to the development of
every increase in the review score of an Airbnb property had a negative Airbnb. We examined the effect across segments and included total
impact of −$25.54 on hotel RevPAR for hotels in the sample. Airbnb inventory and dimensions of customer satisfaction as well as the
In sum, results for H1 support the conclusion for a supplement role rental prices listed on the platform. This allows us to gain depth in the
of Airbnb in the lodging industry. However, results of H2, H3, and H4 analysis on the impact of the platform on the hotel business. We trust
indicate that Airbnb’s effect is more disruptive than it might appear at that such study design sheds light into the complex relationship be-
first. The price of rentals on the platform influence the demand for hotel tween the short terms rental products and services of the peer-to-peer
rooms (H2), which indicate a disruptive effect. And the guest reviews’ platforms and the incumbent’s hotel offer. Indeed, our study reveals
score affects the sales of hotels (H3), which points to a substitution that Airbnb’s offer is not a mere supplement to the market, but rather
effect. In the last steps of the analysis, we explored whether these re- shows substitute characteristics in its long term effects on the patterns
lationships vary across segments of hotel sales.
The results of the mixed model show a significant interaction be- Second, the findings of this study have significant managerial im-
tween the average Airbnb listing price and hotel segments plications. While Airbnb executives claim that their competition with
(F4,80 = 3.02, p = 0.023). In particular, when untangling the interac- traditional hotels is overstated, since it serves different segment of
tion, the analysis revealed Airbnb rental prices had an effect on hotels’ travelers, this study shows that the overlap of customers between
RevPar in the luxury segment (i.e., Segment 5). We observed an in- Airbnb and traditional hotels might be higher than expected. The re-
crease in RevPAR of $0.651 per every dollar increase in AvPrice. The sults show that the effect of Airbnb on hotel sales is intricate. Indeed, it
effect was significantly lower ($0.459) for the upper upscale (segment appears to be based on customers’ dynamic comparison of the price and
4). These results partially support H4 and point to a more complicated value offered by the two products. In other words, our results imply that
relationship between the Airbnb offer and hotel sales performance, guests do consider both products in their decision to purchase and
which we further discuss below. compare the benefits- through the user review rates- and prices of each
against the other. This behavior is indicative of a particular substitution
5. Discussion situation. This study establishes that Airbnb has a disruptive effect since
it has an impact on the growth trajectory of hotel RevPar in the San
While this study contributes to our current knowledge of the effects Francisco market. It therefore, provides empirical evidence for the
of the sharing economy on the performance of incumbents in the hotel ongoing heated debate amongst managers on whether or not Airbnb is
sector, it, nonetheless, has certain limitations. First, our results apply to to disrupt hotel business. On a less positive news for hoteliers, our re-
the San Francisco market and should, therefore, be extended to other sults show that the more Airbnb users are satisfied with the services of
destinations only with caution, if at all. While being the oldest one, and their short rental hosts, the more likely it is for hotel rooms demand to
thus appropriate for a longitudinal study, the market might present decrease. Managers need to, therefore, be aware of the level of service
specificities that might reduce the generalizability of the findings of this and price offered by Airbnb and other sharing platforms in their market
paper. Studies following ours may extend to include more cities and as their customers will be. In sum, hotel managers can no longer ignore
regions for greater generalizability. the Airbnb offer in their destination as the platform offer is not a mere
Secondly, we scraped the data from Airbnb’s site. While scraping the supplement of lodging product.

90
I. Blal et al. International Journal of Hospitality Management 73 (2018) 85–92

Furthermore, our findings on the differences across segments may References

indicate that prospecting guests still perceive that upper-end hotels are
more beneficial. This analysis reveals that for managers of luxury and Allen, D., Berg, C., 2014. The Sharing Economy. How Over-Regulation Could Destroy an
upper scale hotels, the higher the prices of the Airbnb rentals in their Economic Revolution. Institute of Public Affairs, Melbourne (Retrieved January 16,
2018 from https://collaborativeeconomy.com/wp/wp-content/uploads/2015/04/
destination, the higher will be the demand for their room. This is key Allen-D.-and-Berg-C.2014. The-Sharing-Economy.-Institute-of-Public-Affairs.-.pdf).
information when developing their revenue management strategy. BI Intelligence, 2017. Here's-How-Airbnb-Appeals-to-Business-Travelers. (Retrieved on
Since the effect of Airbnb prices on RevPAR is stronger in the luxury January 16, 2018 from http://www.businessinsider.com/heres-how-airbnb-appeals-
to-business-travelers-2017-1.).
segment, it indicates that customers still revert to hotels if they perceive Benner, K., 2017. Inside the Hotel Industry’s Plan to Combat Airbnb. (Retrieved January
the disruptor’s price to be too high. This dynamic reflects a substitution 16, 2018, from https://www.nytimes.com/2017/04/16/technology/inside-the-
effect. As for the results of the differences across segments, they in- hotel-industrys-plan-to-combat-airbnb.html.).
Blal, I., Sturman, M.C., 2014. The differential effects of the quality and quantity of online
dicate that differentiation is key for success in today’s competitive reviews on hotel room sales. Cornell Hosp. Quart. 55 (4), 365–375.
lodging landscape. Bower, J.L., Christensen, C.M., 1995. Disruptive technologies: catching the wave. Harv.
Third, the findings of this paper complement the growing research Bus. Rev. 73 (1), 43–53.
Choi, K.H., Jung, J.H., Ryu, S.Y., Kim, S.D., Yoon, S.M., 2015. The relationship between
on the impact of the sharing economy and our investigation of the
Airbnb and the hotel eevenue: in the case of Korea. Indian J. Sci. Technol. 8 (26).
disruptive effects of new businesses. It reveals that while the sheer Christensen, C.M., 1997. The Innovator's Dilemma: When New Technologies Cause Great
volume of properties listed on the platform does not impact hotel sales, Firms to Fail. Harvard Business School Press, Cambridge, MA.
the value relative to the price offered by the Airbnb product has an Danneels, E., 2004. Disruptive technology reconsidered: a critique and research agenda.
J. Prod. Innov. Manag. 21 (4), 246–258.
effect on lodging performance. Indeed, as predicted by Bower and Dogru, T., Mody, M., Raeisinafchi, C., 2017. Comparing apples and oranges? Examining
Christensen (1995), and tested partially by Guttentag and Smith (2017) the impacts of Airbnb on hotel performance in Boston. Boston Hosp. Rev. 5 (2), 1–15.
from the demand side, the growth, access and wide adoption of Airbnb Edelman, B., Luca, M., Svirsky, D., 2017. Racial discrimination in the sharing economy:
evidence from a field experiment. Am. Econ. J.: Appl. Econ. 9 (2), 1–22.
by a varied swath of customers across several countries − and the Ert, E., Fleischer, A., Magen, N., 2016. Trust and reputation in the sharing economy: the
improvements made by Airbnb to model their services after the tradi- role of personal photos on Airbnb. Tour. Manag. 55, 62–73.
tional hotel services to attract business and mainstream travelers − Gallagher, L., 2017. Q&A with Brian Chesky: Disruption, Leadership, and Airbnb’s Future.
(Retrieved July 23, 2017, from http://fortune.com/2017/03/27/chesky-airbnb-lea-
demonstrates that Airbnb is past the initial stages of disruption. While dership-uber/?utm_source=Triggermail&utm_medium=email&
initially Airbnb appealed mostly to leisure travelers who were attracted utm_campaign=Post%20Blast%20%28bii-e-commerce%29:%20Airbnb%20CEO
to it due to its practical attributes such as cost, location, and amenities %20on%20disrupting%20hotels%20%E2%80%94%20Amazon%20expands
%20plans%20for%20physical%20footprint%20%E2%80%94%20US%20retailers
(Guttentag et al., 2018), it is actively pursuing its reach to business %20have%20lower%20fraud-related%20costs%20than%20foreign%20counterparts
travelers. For example, Airbnb has partnered with Delta Airlines where &utm_term=BII%20List%20E-Comm%20ALL).
airlines guests can earn miles by booking Airbnb properties for ac- Geron, T., 2012. Airbnb Had $56 Million Impact on San Francisco: Study. (Retrieved
January 16, 2018, from https://www.forbes.com/sites/tomiogeron/2012/11/09/
commodation through the airline’s website. Airbnb has piloted a pro-
study-airbnb-had-56-million-impact-on-san-francisco/#7162f9513962).
gram in 2015 called Business Travel Ready, that offers tools for business Geroski, P.A., 1995. Why do we know about entry? Ind. J. Ind. Org. 13 (4), 421–440.
travel managers and identifies properties offering business amenities Gutierrez, J., Garcia-Palomares, J.C., Romanillos, G., Salas-Olmedo, M.H., 2017. The
like Wi-Fi, comfortable laptop friendly workspace, iron, hangers, clean eruption of Airbnb in tourist cities: comparing spatial patterns of hotels and peer-to-
peer accommodation in Barcelona. Tour. Manag. 62, 278–291.
linens etc. such that Airbnb can provide cheaper accommodation that Guttentag, D., 2015. Airbnb: disruptive innovation and the rise of an informal tourism
hotels to business travelers especially in cities with high hotel rates such accommodation sector. Curr. Issues Tour. 18 (12), 1192–1217.
as San Francisco (BI Intelligence, 2017; Kaften, 2017; McDonald, 2017). Guttentag, D., Smith, S., 2017. Assessing Airbnb as a disruptive innovation relative to
hotels: substitution and comparative performance expectations. Int. J. Hosp. Manag.
The disruptive effects of Airbnb could lead to simpler, more con- 64, 1–10.
venient, and price-efficient hospitality experiences. (Guttentag and Guttentag, D., Smith, S., Potwarka, L., Havitz, M., 2018. Why tourists choose Airbnb: a
Smith, 2017). In addition, other spillover effects from increased motivation-based segmentation study. J. Travel Res. 57 (3), 342–359.
Hajibaba, H., Dolnicar, S., 2017. Substitutable by peer-to-peer accommodation networks?
spending in ancillary services like restaurants, etc. will continue to have Ann. Tour. Res. http://dx.doi.org/10.1016/j.annals.2017.05.013.
a growing impact. Based on these theories, we can predict that future Hartmans, A., 2017. Airbnb Now Has More Listings Worldwide than the Top Five Hotel
performance of hotels will depend on their capacity to offer a simpler Brands Combined. (Retrieved January 16, 2018, from http://www.businessinsider.
com/airbnb-total-worldwide-listings-2017-8).
and price-efficient offer to travelers.
Haywood, J., Mayock, P., Freitag, J., Owoo, K.A., Fiorilla, B., 2017. Airbnb & Hotel
The theory development and the empirical analyses in this study, we Performance: An Analysis of Proprietary Data in 13 Global Markets. (Retrieved July
believe, make a contribution to several streams of literature. The study 23, 2017, from http://www.str.com/Media/Default/Research/STR_
AirbnbHotelPerformance.pdf).
tests the effects of disruptive innovation on the trajectory of incum-
Hoffman, L., 2015. Longitudinal Analysis: Modeling Within-person Fluctuation and
bents’ performance. Since Airbnb is still a developing phenomenon in Change. Routledge.
the hospitality industry, which is highly competitive to begin with Airbnb Signs MoU with Maharashtra Government. (Retrieved July 23, 2017, from http://
(Singal, 2015), understanding its long-term effects on RevPAR is critical www.hospitalitybizindia.com/detailNews.aspx?aid=27128&sid=6).
Kaften, C., 2017. With a New Appeal to Business Travelers, Airbnb Aims to Host 100M
for hotel managers. As the disruptor’s expand in the industry, the Guests in 2017. (Retrieved on January, 16, 2018, from https://www.fierceceo.com/
competitive structure of incumbents is very likely to be modified as growth-innovation/a-new-appeal-to-business-travelers-airbnb-aims-to-host-100m-
companies adapt, or fail to adapt, to the disruption. As such, it discusses guests-2017).
Kakar, V., Voelz, J., Wu, J., Franco, J., 2016. The Visible Host: Does Race Guide Airbnb
the competitive dynamics between a new entrant and established in- Rental Rates in San Francisco? (Retrieved January 16, 2018, from https://ssrn.com/
cumbents who analyze a threat and assess their hotel’s ability to react. abstract=2967902).
Finally, it illustrates the supplementary and substitution effects in a co- Kaplan, R.A., Nadler, M.L., 2015. Airbnb: a case study in occupancy regulation and
taxation. Univ. Chicago Law Rev. Online 82 (1), 103–115.
created hospitality service setting where each service encounter is Karlsson, L., Dolnicar, S., 2016. Someone’s been sleeping in my bed. Ann. Tour. Res. 58
variable and likely to be subjectively evaluated. (C), 159–162.
In conclusion, Airbnb is slowly disrupting the lodging industry and Kim, J., Yoon, Y., Zo, H., 2015. Why People Participate in the Sharing Economy: A Social
Exchange Perspective. (Retrieved July 23, 2017, from http://aisel.aisnet.org/
evidence from this study suggests that its effect goes beyond the ‘sup-
pacis2015/76).
plement’ role claimed by its founders and resembles to a substitution Kosová, R., Enz, C.A., 2012. The terrorist attacks of 9/11 and the financial crisis of 2008:
one. Although, our study focused on one market, and generalizability is The impact of external shocks on US hotel performance. Cornell Hosp. Quart. 53 (4),
308–325.
limited, we can reasonably conclude that the impact of the sharing
Lane, J., Woodworth, R.M., 2016. The Sharing Economy Checks In: An Analysis of Airbnb
economy is not related to the volume of the offers of the platform but in the United States. (Retrieved July 23, 2017, from http://www.cbrehotels.com/EN/
rather on the pricing and price-to-value perceived by the guests. Research/Pages/An-Analysis-of-Airbnb-in-the-United-States.aspx).
Lawler, R., 2012. Airbnb: Our Guests Stay Longer and Spend More than Hotel Guests,
Contributing $56M to the San Francisco Economy. (Retrieved July 23, 2017, from

91
I. Blal et al. International Journal of Hospitality Management 73 (2018) 85–92

http://techcrunch.com/2012/11/09/airbnb-research-data-dump/). 23, 2017, from http://www.hotelnewsnow.com/Articles/150801/A-look-at-new-

Markides, C., 2006. Disruptive innovation: in need of better theory. J. Prod. Innov. supply-impact-on-RevPAR-in-US-markets).
Manag. 23, 19–25. Schneiderman, E.T., 2014. Airbnb in the City. (Retrieved July 23, 2017, from https://ag.
McCann, B.T., Vroom, G., 2010. Pricing response to entry and agglomeration effects. ny.gov/pdfs/AIRBNB%20REPORT.pdf).
Strateg. Manage. J. 31 (3), 284–305. Seamans, R., Zhu, F., 2014. Responses to entry in multi-sided markets: the impact of
McDonald, M., 2017. Airbnb’s New Business Travel Program Offers All the Comforts of Craigslist on local newspapers. Manag. Sci. 60 (2), 476–493.
Home. (Retrieved from on January 16, 2018. https://www.tnooz.com/article/ Shankland, S., 2013. Airbnb Founder: Banning Private Rentals as Silly as Banning Cars.
airbnbs-new-business-travel-program-offers-all-the-comfo). Singal, M., 2015. How is the hospitality and tourism industry different? An empirical test
Neeser, D., 2015. Does Airbnb Hurt Hotel Business: Evidence from the Nordic Countries. of some structural characteristics. Int. J. Hosp. Manag. 47, 116–119.
Masters Thesis. Universidad Carlos III de Madrid. Tsai, H., Kang, B., Yeh, R.J., Suh, E., 2006. Examining the hotel room supply and demand
Nickelsberg, M., 2014. Airbnb Partners with the City of Seattle to Provide Lodging During in Las Vegas: a simultaneous equations model. Int. J. Hosp. Manag. 25 (3), 517–524.
Emergencies and Natural Disasters. (Retrieved July 23, 2017, from https://www. Tussyadiah, I., 2015. An exploratory on drivers and deterrents of collaborative con-
geekwire.com/2016/airbnb-partners-city-seattle-provide-lodging-emergencies- sumption in travel. In: Tussyadiah, I., Inversini, A. (Eds.), Information &
natural-disasters/). Communication Technologies in Tourism 2015. Springer International Publishing,
O’Neill, J.W., Ouyang, Y., 2016. From Air Mattresses to Unregulated Business: An Switzerland.
Analysis of the Other Side of Airbnb. AHLA /Penn State Report. (Retrieved on Yglesias, M., 2012. Legalize Airbnb! Slate. (Retrieved July 23, 2017, from http://www.
January 16, 2018 from http://3rxg9qea18zhtl6s2u8jammft-wpengine.netdna-ssl. slate.com/articles/business/moneybox/2012/06/is_airbnb_illegal_why_hotels_are_so_
com/wp-content/uploads/2016/01/0000-0000_PennState_AirBnbReport_011916ee_ upset_about_you_renting_a_bedroom_to_a_stranger_.html).
Embargo.pdf). Zervas, G., Proserpio, D., Byers, J.W., 2015. A First Look at Online Reputation on Airbnb,
O’Neill, J.W., Xiao, Q., 2006. The role of brand affiliation in hotel market value. Cornell Where Every Stay Is Above Average. (Retrieved January 16, 2018, from http://
Hotel Restaur. Adm. Q. 47 (3), 1–14. papers.ssrn.com/sol3/papers.cfm?abstract_id=2554500).
Pizam, A., 2014. Peer-to-peer travel: blessing or blight? Int. J. Hosp. Manag. 38, 118–119. Zervas, G., Proserpio, D., Byers, J.W., 2017. The rise of the sharing economy: estimating
Rael, J., 2017. A Look at New Supply Impact on RevPAR in US Markets. (Retrieved July the impact of Airbnb on the hotel industry. J. Marketing Res. 54 (5), 687–705.

92