Peer-to-peer accommodation prices: City listing concentration and host listing share

Abstract

This article examines the relationship of city listing concentration and host listing share with prices in the peer-to-peer accommodation market. To this end, a multilevel analysis is applied to data from 216,960 Airbnb listings in 45 cities. The results reveal that while city-level concentration does not significantly affect prices, the listing share of each supplier does have a significant effect. This highlights that host listing shares can explain the ability of agents to influence the price of their accommodations even in relatively competitive markets such as peer-to-peer accommodation. Our findings have theoretical implications relating to the idiosyncratic aspects that affect business pricing, as well as practical implications in the private sphere for the rest of the hospitality and tourism industry, and for public policy in terms of regulations.

Keywords

peer-to-peer accommodation Airbnb prices concentration listing share

Introduction

The significance of Airbnb for the short-term accommodation market is evident from the company’s exponential growth since its foundation in 2008 and its undisputed leadership among platforms for renting private homes in the short term (Hajibaba and Dolnicar, 2018). In 2017, Airbnb already surpassed the top five hotel brands combined in terms of listings worldwide (Hartmans, 2017) and in 2021 accounted for more than 7 million listings distributed among 100,000 cities in practically all the countries of the world (Deane, 2021). Considering the length of stay in terms of booked nights, Airbnb outpaced its online travel competitors Booking and Expedia for the entire period for which data are available, from August 2018 to August 2020. Furthermore, Airbnb has the most listing exclusivity: while some listings appear in all or two of these platforms, Airbnb has the highest ratio of listings that appear only on its platform (Transparent, 2020). Moreover, Airbnb exclusively hosts the majority of listings compared to its online vacation rental competitor Vrbo in almost all cities of the world, and the percentage of listings that appear on both Vrbo and Airbnb is typically larger than those that only appear on Vrbo (AirDNA, 2021). Thus, despite the COVID-19 crisis, forecasts point to the continued growth of Airbnb in the coming years (Gassmann et al., 2021).

The Airbnb platform can therefore be seen as a market in itself that numerous hotel companies have been joining (Airbnb, 2021a). A significant feature of this market is that an increasing number of multi-unit hosts are listed (Demir and Emekli, 2021; Dogru et al., 2020a), conventionally referred to in the literature as professionals, which are often specialized companies linked to large hotel groups (Gil and Sequera, 2020). In this vein, some authors have argued that there are no substantial differences between hotels and professional hosts that supply individual rooms within a single building (Dogru et al., 2020a). For this reason, it is tempting to analyze Airbnb from the perspective of an industry, which the US Census defines as “grouping establishments that produce physically similar products by similar processes” (Peltzman, 2014).

Knowing how this competitive environment influences pricing is of primary importance, as price advantage is one of the most informally cited reasons explaining the existence of the peer-to-peer accommodation market, and several studies confirm its significant importance in guests' decisions (So et al., 2018; Tussyadiah, 2016). Although multiple studies have tried to unveil what is behind the price determination tool of Airbnb accommodation, covering a wide range of variables such as physical or locational ones, booking settings or host properties (see Sainaghi, 2020b, for a survey), none so far included the structure of the industry.

This study addresses the relationship of city listing concentration and host listing share with prices in the peer-to-peer market for accommodation. In particular, on the one hand, we analyze how the overall structure of concentration in a city affects prices, and on the other hand, the extent to which host listing accumulation is significant in terms of pricing.

This study’s novelty is that it adds the industrial economics dimension of city listing concentration and host listing accumulation to the peer-to-peer accommodation market, contributing to the literature in several ways. First, the present work proposes theoretical hypotheses to explain the potential effects of city listing concentration and host listing share on prices. Second, our approach empirically tackles both aspects jointly by using data covering 216,960 listings in 45 cities of Europe, the United States, Canada, and Australia, highlighting the significant influence that host listing shares exert on room prices. Third, our empirical findings can be used by hosts to increase their profits without resorting to intensive renting, whereas in terms of public policy, the regulation of the increasing number of hosts with multiple listings can be justified to avoid undesirable consequences in pricing.

Literature review

Airbnb is a peer-to-peer marketplace where supply and demand meet (Dolnicar, 2017). The magnitude of this phenomenon and its relevance for tourism in general and for the hospitality industry in particular have attracted the attention of academic research and generated an abundant literature on the matter, to the point of justifying several surveys (Dann et al., 2019; Dolnicar, 2019; Guttentag, 2019; Prayag and Ozanne, 2018; Sainaghi, 2020a). The Airbnb platform includes individual hosts who offer their homes for economic reasons as much as for non-monetary ones (Camilleri and Neuhofer, 2017; Möhlmann, 2015; Tussyadiah, 2016; Tussyadiah and Pesonen, 2016; So et al., 2018), alongside professional hosts or companies that basically seek economic profit. Professional hosts devise strategies to compete with each other for guests seeking accommodation within the platform. In turn, their behavior also affects the entire hospitality industry, to the extent that the accommodations provided by hotels and those marketed through Airbnb can be considered substitutes (Benítez-Aurioles, 2019; Zervas et al., 2017). On the other hand, Airbnb’s presence can have positive spillovers on the other subsectors of the tourism industry (Basuroy et al., 2020; Dogru et al., 2020b). In the above context, pricing strategy is a basic tool that defines hosts’ market position.

Pricing in the peer-to-peer or Airbnb accommodation market is a salient issue when assessing its effect on the traditional industry (e.g., Dogru et al., 2021). In principle, Airbnb (2021b) provides hosts with a tool (smart pricing) that enables dynamic pricing and the automatic updating of rates according to a set of configuration options. In addition, specialized companies (Wheelhouse, Beyond Pricing, Pricelabs, among others) offer services using different algorithms and data mining to propose effective accommodation prices.

Attempts have been made in the literature to explain accommodation prices in the peer-to-peer market using hedonic regression models where the characteristics of the accommodation, the host or the destination itself are the explanatory variables (Sainaghi, 2020b). Among these, we often find a dichotomous variable such as professional status (1 if the host has more than 1 listing; 0 if just 1) or, more directly, the number of listings offered by each host (“host listing count”), which is linearly related to the listing share in the case of a single city.

Numerous studies have included a dummy for professional hosts as an explanatory variable, albeit with mixed results. In some cases, being a professional host is associated with setting higher prices (Dogru and Pekin, 2017; Magno et al., 2018; Falk et al., 2019; Lorde et al. 2019), while in others it is associated with lowering them (Teubner et al., 2017). In some estimations, the coefficients have not proven to be statistically significant (Gibbs et al., 2018; Li et al., 2019). This diversity of results might be related to other factors, not just to the different empirical strategies used, but to the competitive environment in which each market exists.

Regarding the second measure, that is number of listings, Wang and Nicolau (2017) found a positive link between the number of listings and price that is weaker for higher-priced listings, while Cai et al. (2019) reported an inverted relationship. The latter authors attempted to explain the negative effect of the listing count on prices by multi-listing hosts’ trade-off between booking opportunity and price and by the fiercely competitive environment for multi-listing Airbnb hosts in the area they take as a reference (Hong Kong) in comparison to other markets. Another argument that could help explain the fact that multi-listing hosts offer lower-priced accommodation can be found in the work of Xie and Mao (2017), which hints at a trade-off between host quality and the quantity of listings.

To the best of our knowledge, however, none of these studies considers the composition of production in terms of the percentage offered by each firm in the market. For the traditional accommodation industry, there are several studies on the effect of market share. For example, Abrate et al. (2012) used a sample of 755 hotels in different European cities and observed that prices decrease with competitors with available rooms. Becerra et al. (2013) analyzed data from 1490 Spanish hotels and concluded that competition between hotels favors price decreases among those of lower category and those not owned by chains.

Works relating market share to profitability, such as Buzzell et al. (1975), have explained the link with the attainment of economies of scale that many anti-trust economists, however, believe are unimportant compared to the power gained to bargain more efficiently and end up charging higher prices for a product, such as in Bain (1951). Duverger (2013), who studied the effects of user-generated content on hotel sales, noted, for example, that “a market in which only one hotel is present might behave differently than a market in which 20 hotels are competing,” and that in the first case, the manager might take advantage of the competitive setting to raise the price. In the field of second-home rentals, Saló and Garriga (2011) found that accommodation with the same characteristics is 28.7% more expensive when booked through a high-market-share wholesaler than when booked through intermediaries via the Internet.

To our knowledge, such studies do not deal with the peer-to-peer market for tourist accommodation, while peer-to-peer accommodation papers on the impact of the number of listings or professionalism on prices do not integrate industry concentration yet. We would like to fill this gap in the literature by studying both city listing concentration and host listing share, in an industry where the goods are differentiated and the number of competitors is typically high to see whether the usual results apply or not.

Theoretical framework

The following model is based on Sen (2005), which is an extension of Church and Ware (2000) for more than one dominant firm. Let us assume there are n dominant firms that solve profit maximization (not taking price as fixed) because they have market power, resulting from being more efficient or having a scarce or superior quality product. There are also k fringe competitive firms that act as price-takers and whose total supply corresponds to horizontally summed marginal cost curves. Each firm faces demand functions that are additively separable in quantities produced by firms. If $c_{i}$ is the marginal cost of output facing dominant firm i (not necessarily constant), $s_{i}^{D}$ is the market share of dominant firm i, $ε_{M i}$ is the elasticity of market demand with respect to firm i’s price, $ε_{m i}^{D} > 0$ is the cross price-elasticity of demand of dominant firm’s m product with respect to i’s price, $ε_{j i}^{f}$ is the elasticity of supply of the fringe, $s_{m}^{D}$ is the market share of dominant firms other than firm i, and $s_{j i}^{f}$ is the market share of fringe firm j, and assuming that elasticities are symmetric ( $ε_{m i}^{D}$ = $ε_{m}^{D}$ ∀ m, $ε_{j i}^{f}$ = $ε_{j}^{f}$ ∀ j, $ε_{M i}$ = $ε_{M}$ ∀ i) and all dominant firms charge the same price (p = p₁ = p₂ = … = p_n), then a modified Lerner index can be written such that

\frac{p - c_{i}}{p} = \frac{s_{i}^{D}}{ε_{M} + ε_{j}^{f} \sum_{j = 1}^{k} s_{j}^{f} + ε_{m}^{D} \sum_{m = 1}^{n - 1} s_{m}^{D}}

which shows us the price margin is increasing with the firm’s share. Furthermore, summing both sides across all i’s and multiplying by

Σ_{i = 1}^{n} s_{i}^{D}

we get

\frac{\sum_{i = 1}^{n} s_{i}^{D} (p - c_{i})}{p} = \frac{H H I^{D}}{ε_{M} + ε_{j}^{f} \sum_{j = 1}^{k} s_{j}^{f} + ε_{m}^{D} \sum_{m = 1}^{n - 1} s_{m}^{D}}

\frac{p - c_{i}}{p} = \frac{H H I^{D}}{\sum_{i = 1}^{n} s_{i}^{D} [ε_{M} + (ε_{j}^{f} - ε_{m}^{D}) \sum_{j = 1}^{k} s_{j}^{f} + ε_{m}^{D} (1 - s_{i}^{D})]}

where

H H I^{D} = {\sum_{i = 1}^{n} (s_{i}^{D})}^{2}

is the Hirschman–Herfindahl Index for dominant firms. The implications of increasing

s_{i}^{D}

are the following: the price margin will increase if the share expands at the expense of other dominant firms; but may have an ambiguous effect if taken away from fringe firms. The effect of

H H I^{D}

on price is positive.

We are interested in verifying these relationships in an environment where goods are not perfectly homogeneous, but are to a certain extent substitutable (accommodation). In line with this reasoning, we propose the following hypotheses: 1) The concentration of listings at city level has a significant and positive effect on prices; 2) As the host’s listing share increases, the listing price rises.

Regarding the first hypothesis, we can argue that although the $H H I^{D}$ effect appears to be clear (positive), in practice it is possible to vary shares without changing the $H H I^{D}$ but not the opposite. Thus, the effect of an increase of the $H H I^{D}$ is conditioned by the accompanying movement of shares, and is therefore ambiguous.

Concerning the second hypothesis, although the host listing share can be initially increased through an expansion combined with a reduction of the unit cost and hence of prices to expel competitors (negative correlation between share and price), once they settle into a dominant position, firms tend to adopt monopolistic behaviors, including a raise in their price (positive correlation). Previous research has suggested that hosts usually have low levels of experience in marketing and guest management (Liang et al., 2017, 2018). However, we have to distinguish between professional hosts (with higher listing shares), and non-professional hosts (Chen and Xie, 2017). Benítez-Aurioles (2018) concluded that professional hosts tend to set prices more efficiently compared with non-professional hosts because they are more competent in pricing strategies. Moreover, Wang and Nicolau (2017) found a positive relationship between the number of listings per host and the prices. Other sources of market power, which cause both higher shares and higher prices, are vertical or horizontal differentiation, inelastic demand, entry and exit barriers, and are all present in Airbnb. In particular, the fact that a property is hard to acquire and impossible to identically replicate (not even sales of particular nights in the property), it seems natural to consider that market power or owning a larger number of Airbnb listings causes higher prices, and not that such dominance is caused by pricing choices.

Furthermore, if the host listing share matters more than the market structure in our empirical estimate, we can infer that price discriminating abilities (i.e., the absence of a unique market price even after controlling for the typical differentiating characteristics) are strong. If market structure matters more than host listing share, then the lack of competition rather than being able to deviate from the equilibrium market price will determine whether a markup is charged. As stated above, our hypotheses are that both the concentration of listings at city level and host listing share will have a positive influence on price; which is stronger will depend on which of the two theoretical hypotheses prevails in practice.

Data and empirical approach

Data

Our data come from Insideairbnb.com, a website that is not associated with Airbnb or its competitors (Insideairbnb, 2019). A substantial number of empirical works on Airbnb have used this resource (Wang and Nicolau, 2017; Benítez-Aurioles, 2018; Chattopadhyay and Mitra, 2020; etc.). In a recent literature review by Dann et al. (2019), the authors observed the growing use of Insideairbnb.com as a data repository and suggested that future research performed on this basis will allow better result comparability. We also obtain regional GDP per capita data (thousands of euros) from Eurostat (2019) and the UK’s Office for National Statistics (ONS, 2019) to control for differences in the average cost of living that vary the price of accommodation.

Table 1 shows the cities (or level 3 in our multilevel analysis) for which InsideAirbnb performed information scraping on ads for accommodation in the months of July 2018 and July 2019, with the exact days of collection, number of listings (level 1) and hosts (level 2). The database contains 216,960 observations corresponding to 45 cities in Europe, the United States, Canada and Australia. To obtain more homogenous units for comparison, we have focused our analysis on the most central, touristic and Airbnb-dense neighborhoods of the selected cities. Although all the selected cities have a certain tourism demand, some outermost areas might be outside the geographical boundaries of the main Airbnb market (e.g., Inner vs. Outer London). After observing the distribution of listings among the defined neighborhoods and districts of the cities, we observed the following recurrent fact that can be used as a rule-of-thumb for our selection: 50% of the listings are concentrated in the central, Airbnb-dense neighborhoods, while the rest of the listings are scattered across the remaining less-demanded neighborhoods.

Table 1.

Dataset details.

City no.	City (L.3)	Day July 18–19	Listings (L.1)	Hosts (L.2)	HHI × 10,000
1	Amsterdam	6–4	6731	6208	2.135
2	Antwerp	28–25	621	485	32.854
3	Athens	16–10	2201	1653	12.549
4	Austin	10–8	3394	2632	12.551
5	Barcelona	10–7	4844	2996	22.998
6	Berlin	31–8	9664	8783	1.366
7	Bologna	25–24	1121	785	23.101
8	Bordeaux	18–14	3563	2948	7.405
9	Boston	18–14	979	611	151.757
10	Bristol	11–10	600	484	26.778
11	Brussels	18–14	2162	1640	21.886
12	Chicago	17–14	1442	1031	32.587
13	Columbus	29–27	200	140	102.5
14	Copenhagen	16-17	8575	8165	1.355
15	D.C. Washington	31–29	3699	2672	9.1
16	Denver	28–25	657	564	24.904
17	Dublin	28–25	3680	2804	7.572
18	Edinburgh	28–25	3128	2462	8.894
19	Geneva	22–16	1059	896	17.914
20	Gent	31–29	844	678	20.833
21	Lisbon	29–26	3338	1752	22.907
22	London	7–5	25,073	19,693	1.544
23	Los Angeles	5–4	16,480	11,085	2.025
24	Lyon	25–24	3624	3367	3.408
25	Madrid	10–7	8047	4594	11.767
26	Malaga	31–30	1450	763	73.76
27	Manchester	17–8	575	403	44.31
28	Milan	11–8	4352	3537	6.104
29	Montreal	11–10	5355	4229	13.905
30	Naples	31–8	3771	2347	7.204
31	Nashville	13–11	2706	1307	102.638
32	New Orleans	5–3	2391	1235	69.039
33	New York	5–2	25,131	21,685	0.628
34	Paris	8–5	19,652	18,368	0.866
35	Portland	10–7	782	676	18.151
36	Quebec	9–6	671	487	42.799
37	Rome	11–8	6902	3722	6.866
38	San Diego	16–13	3156	1960	35.997
39	San Francisco	5–2	1160	808	38.065
40	Seattle	17–13	1690	1242	28.024
41	Sydney	6–4	9894	8550	1.835
42	Toronto	6–4	3853	3101	5.34
43	Vancouver	10–7	1760	1393	9.465
44	Venice	9–7	3001	1631	22.093
45	Vienna	11–11	2982	2247	11.626
TOTAL			216,960	168,819

We consider different characteristics of the listings that are described in the Empirical Strategy section. Additionally, we use the Herfindahl–Hirschman index (HHI) as a measure of the concentration of listings at the city level. Since its proposal in the mid-20th century, the HHI has been used time and again in the field of industrial economics, even by anti-trust agencies, to inform decisions on takeovers or mergers (Roberts, 2014). The HHI has been deemed superior to other indexes, such as the sum of the top four market shares (Kwoka, 1981; Schmalensee, 1977).

When it comes to tourism research, the HHI has been employed to measure the degree of concentration of tourism activity, both internationally (Croes and Kubickova, 2013; Liu et al., 2018) and regionally (Fernandes et al., 2020; Majewska, 2015). The index has also been utilized to calculate the concentration of revenue among hotels (Davies, 1999; Pan, 2005). We include information on the HHI index by city in the last column of Table 1. Considering a host as a firm, a value of 1 divided by the number of listings in the city (or multiplied by 10,000, as is often done in the literature) implies a perfectly equal spread, while 1 (or 10,000) implies a monopoly. The US Department of Justice and Federal Trade Commission considers an industry with an HHI of 1500–1800 to be moderately concentrated and an HHI above 1800 to represent a high concentration (Peltzman, 2014). The lowest HHI are found in the large metropolises of New York, Paris, Copenhagen, Berlin, London and Sydney (all below 2), while the highest are found in some smaller American cities like Columbus, Nashville or Boston (100 to 152). This is related to an intrinsic feature of the HHI: while the maximum is always 10,000, the possible minima are 10,000 divided by the number of producers. Hence, we cannot expect very high indices given the large number of hosts accounted for, even in small cities. In fact, our values are quite significant compared to their lowest theoretical values (1/hosts). Moreover, we have computed other concentration indexes and the values are around those considered high in other areas of the economic literature (i.e., 0.3 for the Gini income distribution index).

There are alternative indexes to the HHI. One example is the Lerner (1934) index: (P–C)/P, where P is the price and C the marginal cost, which in the literature has been found to be positively related to several proxies of market concentration. Theoretically, there are still more robust indexes than the price-cost margin as a measure of competition, such as the Boone (2008) index, which measures the elasticity of firms’ profits or market shares to their market inefficiency, assuming that firms are punished for being inefficient. Unfortunately, although we have ways to estimate the revenue (price × reviews), we do not have any information on the marginal cost, which could help us construct these measures. Nevertheless, we were able to calculate alternative concentration indexes, such as the S-concentration ratio (the sum of the listing shares of the S top hosts in the city) and other concentration indexes in the income distribution literature, such as the Gini index and Entropy measures.

Empirical strategy

Our empirical work tests whether a higher concentration at city level and a higher host listing share are associated with higher prices. Specifically, we test for the host share (second-level variable) and city concentration (third-level variable) taking into account that listings are the basic (first) level. The hierarchical structure of the data makes it advisable to apply multilevel techniques, as they correct the estimation of standard errors by controlling for the correlation at superior levels, since listings by the same host or hosts from the same city may be more correlated than listings of a different host or different hosts from different cities. For our purpose, these techniques are preferred to ordinary least squares (OLS) because the downward bias of OLS standard errors when aggregate level data is combined with micro level data leads the analyst to reject the null hypothesis of no effect (Moulton, 1990). Moreover, when OLS is corrected for clustering, the null hypothesis is over-rejected compared to the estimates obtained from multilevel models (Cheah, 2009; King and Roberts, 2015). Additionally, multilevel models can account for a city’s unmeasured features (i.e., legislation, events, cultural norms, type of tourism), as well as host characteristics (i.e., main occupation, socio-economic background, motivations) that might affect pricing, and the relationship between listing ownership and price. This is what the multilevel models also intend to control for, like any other city-level unobservable or not (fully) measurable idiosyncrasy.

Our specification, therefore, is a three-level model with random slope for the host listing share and the listing price in logarithmic form as a dependent variable

\log p r i c e_{i h c} = β_{0} + β_{1} X_{i h c} + β_{2} Y_{h c} + β_{3} s h a r e_{h c} + β_{4} Z_{c} + β_{5} H H I_{c} + θ_{0 c} + θ_{1 h c} + θ_{2 c} s h a r e_{h c} + u_{i h c}

(1)

where

\log p r i c e_{i h c}

is the logarithm of the price of listing

i

that belongs to host

h

and city

c

X_{i h c}

are variables at lower level (listing-level),

Y_{h c}

are host-level variables,

s h a r e_{h c}

is the host listing share,

Z_{c}

is a variable at higher level (city-level), and

H H I_{c}

is the HHI index in the city.

θ_{0 c}

is the effect of city

c

on price,

θ_{1 h}

is the effect of host

h

on price,

θ_{2 c} s h a r e_{h c}

allows the coefficient for the host listing share to vary across cities and

u_{i h c}

is the listing-level error.

Table 2 shows the breakdown of variables used in (1). Listing- and host-level regressors were included following previous work (Benítez-Aurioles, 2018; Chattopadhyay and Mitra, 2020; Wang and Nicolau, 2017).

Table 2.

Descriptive statistics.

Variables	Description	Mean	SD
Listing level
Log price	Nightly log price (in $)	4.781	0.789
Bedrooms	Number of bedrooms	1.356	0.908
Bathrooms	Number of bathrooms	1.231	0.593
Entire	1 if accommodation is entire home, 0 if shared room^a	0.684	0.465
Private	1 if accommodation is private room, 0 if shared room^a	0.305	0.460
Distance	Distance in km to the city hall	4.579	4.917
L. no. reviews	Number of reviews received 1 year ago (2018)	23.892	45.022
Flexible	1 if cancellation policy is flexible, 0 if stricter^a	0.306	0.461
Instant	1 if instant booking is enabled, 0 otherwise^a	0.391	0.488
Breakfast	1 if breakfast included, 0 otherwise^a	0.116	0.320
Family/kid	1 if family-friendly, 0 otherwise^a	0.495	0.500
Smoking	1 if smoking allowed, 0 otherwise^a	0.114	0.318
Host level
Superhost	1 if owner is superhost, 0 otherwise	0.223	0.416
Listing share	Host’s total listings into total number of listings in the city	0.100	0.005
City level
HHI	Herfindahl–Hirschman index	0.001	0.002
N. Listings	Total number of listings in the city in thousands	11.959	8.916
GDP	GDP per capita of the city’s region in 2018 in thousands of euros	56.405	22.285
Restrictions	1 if city has major laws regulating short-term rentals (restriction to primary residences, yearly renting time limit, registry, fees…) at the time of data collection: Amsterdam, Barcelona, Boston, London Los Angeles, New Orleans, New York, Paris, San Diego, San Francisco, Seattle, and Vancouver.

^aThe standard deviation for dummy variables is $\sqrt{μ (1 - μ)}$ where $μ$ is the mean of the variable or proportion of observations coded as 1.

For listing-level variables, we use the number of bedrooms as a set of 7 dummy variables: having 0 (studio) bedroom, 1 bedroom, … 6 bedrooms or more. Hence, we do not assume, for instance, that the effects of moving from 1 to 2 bedrooms are similar to the effects of moving from 2 to 3 bedrooms. Other listing-level variables are the number of bathrooms and whether an entire home or a private room is listed (taking shared rooms as the base category). The empirical evidence also suggests the use of distance to the city center (Benítez-Aurioles, 2018), which is conventionally located at the city hall. The number of reviews is an indication of previous demand and is a variable that is also likely to influence price. We control for the level of reviews in the preceding year (2018) to reduce possible endogeneity between the present number of reviews and current prices. Regarding booking policy, two indicators are used that take the value of 1 when flexible cancellation (full refund of the booking if canceled at least 24 h in advance) and instant booking are available, respectively. Among the amenities, we have selected breakfast, family-friendly and smoking allowed.

For the host-level variables, we incorporate superhost status, which refers to hosts that have met a certain performance threshold established by Airbnb (2021c). Furthermore, we include Listing share, which is defined as the host’s listing share in their city. That is, for each host h

L i s t i n g s h a r e_{h, c} = \frac{l_{h, c}}{L_{c}}

where

l_{h c}

is the number of listings owned per host h of city c and

L_{c}

is the number of listings in city c.

As for city-level variables, we include GDP per capita (GDP) as it might influence the marginal cost of offering the accommodation since it is related to the cost of living, as proposed in the theoretical section. We also include city size in terms of thousands of listings, $L_{c}$ , to account for the possible effects on price due to the varying importance of inhabitants, area and tourism of different cities. Concerning the city concentration variable, we use HHI, which is calculated as follows

H H I_{c} = \sum_{h = 1}^{H_{c}} L i s t i n g s h a r e_{h, c}^{2}

where

H_{c}

is the total number of hosts in city

c

. ranges from

1 / H_{c}

to 1, where the minimum value corresponds to perfect competition and the maximum value to monopoly. As a robustness check for this variable, we apply alternative city concentration measures: the S-concentration ratio (

C R_{c}^{S}

, shortened to CRS), which is the sum of the shares of the S top hosts in the city

C R_{c}^{S} = \sum_{i = 1}^{S} L i s t i n g s h a r e_{h, c} .

As an alternative to the host listing share, we use the variable crs that equals 1 if the host is among the S top hosts in the city and 0 otherwise. We run several tests using values of S from 2 to 8. We also perform the estimation for alternative city concentration measures, such as the Gini index and Entropy measures for parameter 0, 1 (Theil index) and 2 (see the Online Appendix for the expression of these measures).

Results

We use a bottom-up estimation strategy that consists of adding parameters in a step-by-step fashion. To start with, we run a regression (Model 1) with all listing- and host-level variables (except the host listing share) and HHI index. From there, other city level variables are added followed by the host listing share (Model 2), which includes a random slope that allows the effect of listing accumulation at the host level to vary across cities.

A likelihood ratio test comparing the three-level model to the single level one concludes that the three-level model fits the data significantly better

(χ_{2}^{2} = 107806.53, p = 0.000)

. We also run two likelihood ratio tests to compare the simpler two-level listings-within-hosts model

(χ_{1}^{2} = 44802.88, p = 0.000)

and to the two-level listings-within-city model

(χ_{1}^{2} = 54419.94, p = 0.000)

to confirm that both the host- and the city-level variances are separately significant. Hence, we conclude that, in terms of price, listings of the same host are significantly closer than listings of different hosts. Similarly, listings from the same city are significantly more homogenous than listings from different cities. The estimated parameters are fairly stable across alternative specifications, thus suggesting the robustness of the estimation procedure. The results of our estimates are presented in Table 3.

Table 3.

Multilevel regression of prices on listing-, host-, and city-level variables.

	1	2
Bedrooms 0 (Studio, ref.)
1	0.135***	0.135***
1	[0.004]	[0.004]
2	0.392***	0.392***
2	[0.004]	[0.004]
3	0.655***	0.655***
3	[0.005]	[0.005]
4	0.890***	0.890***
4	[0.007]	[0.007]
5	1.067***	1.067***
5	[0.012]	[0.012]
6 or more	1.154***	1.155***
6 or more	[0.017]	[0.017]
Bathrooms	0.133***	0.133***
Bathrooms	[0.002]	[0.002]
Entire	0.974***	0.974***
Entire	[0.010]	[0.010]
Private	0.381***	0.381***
Private	[0.010]	[0.010]
Distance to center	−0.011***	−0.011***
Distance to center	[0.000]	[0.000]
L. no. reviews	−0.001***	−0.001***
L. no. reviews	[0.000]	[0.000]
Flexible	−0.142***	−0.142***
Flexible	[0.002]	[0.002]
Instant booking	−0.013***	−0.013***
Instant booking	[0.002]	[0.002]
Breakfast	0.057***	0.057***
Breakfast	[0.003]	[0.003]
Family/kid	0.082***	0.082***
Family/kid	[0.002]	[0.002]
Smoking	−0.107***	−0.107***
Smoking	[0.003]	[0.003]
Superhost	0.076***	0.076***
Superhost	[0.003]	[0.003]
HHI	13.761	24.825
HHI	[17.853]	[18.118]
No. Listings		0.026***
No. Listings		[0.009]
GDP		0.001
GDP		[0.001]
Listing share		79.285***
Listing share		[21.974]
Constant	3.579***	3.295***
Observations	216,960	216,960
Number of groups	45	45

Significant at ***0.01 level; **0.05 level; *0.10 level. Standard errors in parentheses.

In all models, the variables that have a positive influence are number of bedrooms, bathrooms, being an entire home or a private room (as compared to a shared room), serving free breakfast, being family-friendly and having a superhost. The variables with a negative association are distance to the city center, number of reviews, flexible booking policy and smoking being allowed.

Each extra bedroom does not have a constant price effect as suggested by the coefficient for the number of bedrooms (for 0 to 1 room the price increases by 13%, from 0 to 2 by 39% and so on). Each extra bathroom increases the price by 13%. Entire homes are 97% more expensive than shared rooms and private rooms are 38% more expensive than shared rooms. Each km of distance to the center reduces the price by 1.1%. For each additional review, the accommodation is typically 0.1% less expensive. Accommodations allowing flexible cancellation are on average 14% cheaper and instant booking is 1% less expensive. Serving free breakfast raises the price by 5.7%, while being a family/kid-friendly type of accommodation does so by 8%. Smoking allowed reduces the price of the accommodation by 11%, while being a superhost increases the price by 7.6%.

In respect to the concentration of listings at the city level (HHI), it does not appear to be significant. This empirical result helps resolve the ambiguity of the theoretical model: accepting that HHI movements cannot be fully disentangled from share movements, we find that the HHI has a neither positive nor negative effects, perhaps due to this market being weakly concentrated to start with, and hence the variations of the HHI caused by low share values being insignificant on prices.

This is also robust to the inclusion of other city characteristics and the host listing share (Model 2). In Model 2, the total number of listings in the city (i.e., its dimension in terms of Airbnb) is relevant, as the price increases by 2.6% for each increase in 1000 listings, while GDP seems to exert a non-significant effect on price. Additionally, an increase in a host’s listing share by 1% point increases the price by 0.8%. Thus, listings of hosts with greater listing shares are associated to higher prices, possibly due to their experience of operating multiple units, confirming the second hypothesis. At the same time, the effect of an increase in the concentration of listings at the city level (HHI) remains not statistically significant, as in model 1.

For an evaluation of the relative importance of the variables, we compute the standardized coefficients of all variables in the regression (see Table A1 in the online Appendix). We observe that, while the HHI index is not statistically significant, an increase in the listing share of one standard deviation implies an increase in the price of around 0.5%.

Figure 1 shows the city-level effects of the host Listing share on prices within a 95% confidence interval. This reveals an interesting fact that was missing at the aggregate level: there are only three cities¹ (San Francisco, San Diego and Austin) whose listing share effect on prices differs remarkably from the rest. In fact, the host listing share effect is below the average effect in half of the cities (23 out of 45).

To sum up, our main results show that while the rise in the level of concentration of listings at the city level does not have a significant effect, increasing the host Listing share does indeed rise the level of prices (in particular, 1 additional percentage point of a listing share would allow increasing the price by 0.8%), even though this impact varies across cities.

Robustness analysis

To assess the robustness of our results, we introduce several variations in the specification of Model (2). Specifically, we have checked the robustness of the results under different definitions of city concentration: when CRS is chosen as the concentration index at the city level for different values of S from 2 to 8, or when the Gini or Entropy measures of concentration are introduced instead of the HHI index, the effect is insignificant (Table A3 of the Online Appendix). Therefore, the concentration of listings at the city level—irrespectively of the index used—has no significant influence on prices. However, the accumulation of listings per host is significant for different values of S from crs2 to crs8: an increase of the top 2 (8) hosts in a city increases the price by around 27% (21%). In the same line, a one percentage point increase in host listing share is significant for the Gini index and Entropy measures models with parameter $w = 0, 1, 2$ (around 0.8% increase).

Airbnb properties are very heterogeneous. Rooms in shared flats may not be competitors for whole flats. This has been accounted for in a robustness check in which we have performed separate regressions for entire and non-entire flats. Table A4 in the Online Appendix shows that the results remain unchanged.

Finally, restrictions on Airbnb in certain cities could be an issue, given that some cities have enacted stricter public regulations of the peer-to-peer accommodation market than others. In Barcelona, for instance, no new licenses are being issued for Airbnb properties. Other cities have imposed quantitative restrictions (e.g., on the number of nights that accommodations can be booked) or qualitative restrictions (e.g., requiring the host’s presence during the rental period in multifamily dwellings) (see Nieuwland and Van Melik, 2020). At the time of our data collection, Paris, Barcelona, Berlin, Amsterdam and London are some of the cities that had already begun to enforce laws regulating the presence of Airbnb.

Accordingly, such regulations could modulate the relationship between the HHI and prices. In this sense, we have performed alternative specifications including a dummy containing information on cities with restrictions, as defined by variable Restrictions in Table 2. The results are reported in Table A5 in the Appendix and show that, even though prices are generally higher in cities that have imposed restrictions, there is no evidence of a differential effect of concentration at the city level nor of the host listing share on prices for cities with these restrictions.

Discussion and conclusions

The main result of our research is that an increase in the concentration of supply in the peer-to-peer market for accommodation at the city level does not have a statistically significant effect on prices. However, the increase of each supplier’s listing share does influence prices in a statistically significant way, increasing the price per each extra percentage point of listing share by 0.8%. In a nutshell, the influence on prices of the level of concentration in the city is weak, even though the accumulation of listings is relevant for the host, who may charge a higher price. These findings have theoretical and practical implications.

On a theoretical level, it helps understand the price formation process both individually and in the market as a whole. From a conventional point of view, the canonical model of market representation, based on the analysis of supply and demand, predicts that when the product is homogeneous and the number of suppliers is high, the firm becomes a price-taker insofar as it lacks freedom to set the price of its product. Consequently, product differentiation would give firms some control over their prices when they operate in a monopolistically competitive market, even when competing with a multitude of firms, as occurs, for example, in the peer-to-peer market for tourist accommodation. That is, a certain market power is gained through the listing share when there is product differentiation. Perhaps for this reason, following Porter’s seminal contribution (1979), a relevant part of the theoretical research has been oriented toward the competitive advantages granted by product differentiation. In this sense, the existence of a direct relationship between the host listing share and prices, which is diluted when the market as a whole is considered, could help us understand why firms that operate in the peer-to-peer market in particular and in the hospitality industry in general perform differently when operating in similar environments. In addition to product differentiation, listing share, although relatively small in a market with many suppliers, helps to understand the ability of agents to influence the price of their accommodations. This idea reinforces the interest of a line of research on the idiosyncratic aspects that affect business pricing (Hunt and Morgan, 1995).

On a practical level, there are also some implications, both in the private and public spheres. For an inexperienced rental owner, in contrast to the traditional qualified hotel manager, it is necessary to recognize the pricing determinants of Airbnb’s online tourist rentals. Our contribution is to signal the degree of accumulation of listings by hosts as one of these determinants. As the number of listings grows, not only can economies of scale appear due to cost savings, but the incentives for hosts to acquire organizational learning also increase. Hosts need to acknowledge the greater ability to raise prices that is obtained when accumulating more listings, which can increase average profits without the need to rent intensively. Increasing prices, conveniently combined with taking care of a reduced number of listing attributes identified in the literature, can help improve a tourist’s rental experience and offset more expensive prices. Therefore, the power derived from accumulation can help adjust room prices, either by maintaining a base price for a specific period or by setting the price dynamically.

Given the increased professionalization of hosts and the entry of large corporations managing an ample accommodation portfolio, more efficient pricing strategies related to the accumulation of the supply at the host level would be expected. This could be interpreted as a threat to the traditional hospitality industry, as long as visitors see them as highly substitutable. In this sense, in line with what has already been pointed out by Dogru et al. (2021), instead of seeking to attract customers exclusively through price reductions, hotels could perhaps emphasize competitive advantages that cannot be provided by Airbnb accommodations, such as the professional care provided by well-trained employees (Mody et al., 2019).

Although here the listing share takes relatively lower values than in other industries, the negative effects of short-term tourist rentals on long-term local rentals are still a concern (Horn and Merante, 2017). If landlords see the former as more profitable, but higher listing ownership per person allows maintaining the price markup, then affecting the distribution of listings is something that could ease pressure on the local real estate market without necessarily affecting the number of tourists visiting the city. Perhaps, as some cities have already done, attempts should be made to restrict short-term rentals to hosts’ primary residences only; that is, to reduce the number of listing ownerships to 1, in practice. This also has relevant implications in the field of public policy. Depending on the business model to be promoted and the social and economic impacts caused in each case by the development of the peer-to-peer market for tourist accommodation, a system for granting permits to operate in an area which considers each host’s listing share, among other factors, could be suggested (Benítez-Aurioles, 2021).

Our research opens new paths in this field and could be expanded in further directions. First, our study could be complemented with a quantile regression estimation to examine the effect of city level concentration and host listing share not only on the conditional mean prices but also on the conditional median (or other quantiles) prices. This type of regression is useful to predict relationships between variables in cases where there is no relationship or only a weak relationship between the means of such variables. Second, regarding the temporal scope, our data represent a snapshot in time. Complementary conclusions could be derived from a comparative static analysis in which a possible change in the role of concentration is explored over time or at different seasons of the year. Third, a natural question that arises is how the COVID-19 pandemic could have affected our conclusions. If the hypotheses of Dolnicar and Zare (2020) turn out to be true, Airbnb would return to a more classical sharing ethos at the expense of investor-hosted listings, thus potentially evening out the distribution of listings per hosts and reducing the average prices. Additionally, such ‘temporal expansions’ could also be applied in the geographical dimension, such as in the locational concentration analyses of Xie and Mao (2019) and their managerial implications. Finally, the use of alternative market power estimators using an accurate quantification of marginal costs or allowing for other non-linearities in the dependence of the HHI is encouraged.

Supplemental Material

Supplemental Material—Peer-to-peer accommodation prices: City listing concentration and host listing share

Supplemental Material for Peer-to-peer accommodation prices: City listing concentration and host listing share by Elena Bárcena-Martín, Beatriz Benítez-Aurioles, and Salvador Pérez-Moreno in Tourism Economics

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Beatriz Benítez-Aurioles

Supplemental Material

Supplemental material for this article is available online.

Note

Author biographies

Elena Bárcena-Martín, barcenae@uma.es, is Professor of Statistics and Econometrics at the University of Malaga, and Treasurer of the Society for the Study of Economic Inequality. She received her MSc in Econometrics and Mathematical Economics at the London School of Economics and her PhD in Economics from the University of Malaga. She has published in journals such as Socio-Economic Review, Journal of Social Policy, Review of Income and Wealth, and Journal of Economic Inequality.

Beatriz Benítez-Aurioles, bbaurioles@uma.es, is Assistant Professor of Economic Policy at the University of Malaga. MSc Economics and Finance at CEMFI. Expanded studies at London School of Economics and University College of London. One of her research areas consists of the analysis of peer-to-peer market for tourist accommodation, publishing in Current Issues in Tourism, International Journal of Contemporary Hospitality Management, Tourism Management, Tourism Economics, etc.

Salvador Pérez-Moreno, sperezmoreno@uma.es, is Professor of Economic Policy at the University of Malaga, Research Associate of the Cambridge Centre for Economic and Public Policy, University of Cambridge, and Affiliated Researcher of UNU-MERIT, United Nations University and Maastricht University. He has published in journals such as Economic Modelling, European Planning Studies, Journal of Business Research, Journal of Economic Policy Reform, Journal of Social Policy, Review of Income and Wealth, and Social Indicators Research.

References

Abrate

Fraquelli

Viglia

(2012) Dynamic pricing strategies: evidence from European hotels. International Journal of Hospitality Management 31(1): 160–168.

Airbnb (2021a) What are Airbnb’s standards for hotels and other hospitality businesses? Available at: https://www.airbnb.com/help/article/1526/what-are-airbnbs-standards-for-hotels-and-other-hospitality-businesses?locale=en&_set_bev_on_new_domain=1617197869_Y2EyYTQwNjc5MmE1 (Accessed June 23, 2022).

Airbnb (2021b) How do I turn smart pricing on or off? Available at: https://www.airbnb.com/help/article/1168/how-do-i-turn-smart-pricing-on-or-off (Accessed June 23, 2022).

Airbnb (2021c) Superhost: recognising the best in hospitality. Available at: https://www.airbnb.com/superhost (Accessed June 23, 2022).

AirDNA (2021) Market overview. Available at: https://www.airdna.co/vacation-rental-data/ (Accessed 20 February, 2021).

Bain

(1951) Relationship of profit rate to industry concentration in American manufacturing 1936–1940. Quarterly Journal of Economics 65: 293–324.

Basuroy

Kim

Proserpio

(2020) Estimating the impact of Airbnb on the local economy: evidence from the restaurant industry. Available at: https://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID3788302_code3681360.pdf?abstractid=3516983&mirid=1 (Accessed June 23, 2022).

Becerra

Santaló

Silva

(2013) Being better vs. being different: differentiation, competition, and pricing strategies in the Spanish hotel industry. Tourism Management 34: 71–79.

Benítez-Aurioles

(2018) Why are flexible booking policies priced negatively? Tourism Management 67: 312–325.

10.

Benítez-Aurioles

(2019) Is Airbnb bad for hotels? In: Current Issues in Tourism (pp, 1–4). Milton Park, UK: Taylor & Francis Online. DOI: 10.1080/13683500.2019.1646226

11.

Benítez-Aurioles

(2021) A proposal to regulate the peer-to-peer market for tourist accommodation. International Journal of Tourism Research 23(1): 70–78.

12.

Boone

(2008) A new way to measure competition. The Economic Journal 118(531): 1245–1261.

(1975) Market share-a key to profitability. Harvard Business Review 53(1): 97–106.

, et al. (2019) Price determinants of Airbnb listings: evidence from Hong Kong. Tourism Analysis 24(2): 227–242.

15.

Camilleri

Neuhofer

(2017) Value co-creation and co-destruction in the Airbnb sharing economy. International Journal of Contemporary Hospitality Management 29(9): 2322–2340.

16.

Chattopadhyay

Mitra

(2020) What Airbnb host listings influence peer-to-peer tourist accommodation price? Journal of Hospitality & Tourism Research 44(4): 597–623.

17.

Cheah

(2009) Clustering Standard Errors or Modeling Multilevel Data? New York City, NY: University of Columbia, pp. 2–4.

18.

Chen

Xie

(2017) Consumer valuation of Airbnb listings: a hedonic pricing approach, International. Journal of Contemporary Hospitality Management 29(9): 2405–2424.

19.

Croes

Kubickova

(2013) From potential to ability to compete: towards a performance-based tourism competitiveness index. Journal of Destination Marketing & Management 2(3): 146–154.

(2019) Poster child and Guinea pig–insights from a structured literature review on Airbnb. International Journal of Contemporary Hospitality Management 31(1): 427–473.

21.

Davies

(1999) Industrial organization the UK hotel sector. Annals of Tourism Research 26(2): 294–311.

22.

Deane

(2021) 2021 Airbnb Statistics: Usage, Demographics, and Revenue Growth. Available at: https://www.stratosjets.com/blog/airbnb-statistics/ (Accessed June 23, 2022).

23.

Demir

Emekli

(2021) Is Airbnb no longer a sharing economy platform? Evidence from Europe’s top 10 Airbnb destinations. Anatolia 32: 1–19. DOI: 10.1080/13032917.2021.1890626

24.

Dogru

Pekin

(2017) What do guests value most in Airbnb accommodations? An application of the hedonic approach. Boston Hospitality Review 5. Available at: https://vtechworks.lib.vt.edu/bitstream/handle/10919/79602/What-do-guests-value-most-in-Airbnb-accommodations_An-application-of-the-hedonic-pricing-approach_Dogru-Pekin.pdf?sequence=1 (Accessed June 23, 2022).

25.

Dogru

Mody

Line

, et al. (2021) The effect of Airbnb on hotel performance: comparing single-and multi-unit host listings in the United States. Cornell Hospitality Quarterly. Online First. DOI: 10.1177/1938965521993083

, et al. (2020a) Airbnb 2.0: is it a sharing economy platform or a lodging corporation? Tourism Management 78: 104049. DOI: 10.1016/j.tourman.2019.104049

, et al. (2020b) The Airbnb paradox: positive employment effects in the hospitality industry. Tourism Management 77: 104001. DOI: 10.1016/j.tourman.2019.104001

28.

Dolnicar

(2019) A review of research into paid online peer-to-peer accommodation: launching the Annals of Tourism Research curated collection on peer-to-peer accommodation. Annals of Tourism Research 75: 248–264.

29.

Dolnicar

Zare

(2020) COVID19 and Airbnb–disrupting the disruptor. Annals of Tourism Research 83: 102961.

30.

Dolnicar

(2017) Entrepreneurship opportunities. In: Peer-to-peer Accommodation Networks. Oxford, NY: Goodfellow Publishers, pp. 77–86.

31.

Duverger

(2013) Curvilinear effects of user-generated content on hotels’ market share: a dynamic panel-data analysis. Journal of Travel Research 52(4): 465–478.

32.

Eurostat (2019) Regional gross domestic product by NUTS 2 regions - million EUR. Available at: https://ec.europa.eu/eurostat/web/regions/data/main-tables (Accessed June 23, 2022).

33.

Falk

Larpin

Scaglione

(2019) The role of specific attributes in determining prices of Airbnb listings in rural and urban locations. International Journal of Hospitality Management 83: 132–140.

34.

Fernandes

Nunes

Veloso

, et al. (2020) Spatial and temporal concentration of tourism supply and demand in Northern Portugal. Application of the Herfindahl-Hirschman index. In: Advances in Tourism, Technology and Smart Systems. Singapore: Springer, pp. 263–273.

35.

Gassmann

S-E

Nunkoo

Tiberius

, et al. (2021) My home is your castle: forecasting the future of accommodation sharing. International Journal of Contemporary Hospitality Management 33(2): 467–489.

36.

Gibbs

Guttentag

Gretzel

, et al. (2018) Use of dynamic pricing strategies by Airbnb hosts. International Journal of Contemporary Hospitality Management 30(1): 2–20.

37.

Gil

Sequera

(2020) The professionalization of Airbnb in Madrid: far from a collaborative economy. Current Issues in Tourism. Online First. DOI: 10.1080/13683500.2020.1757628

38.

Guttentag

(2019) Progress on Airbnb: a literature review. Journal of Hospitality and Tourism Technology 10(4): 814–844.

39.

Hajibaba

Dolnicar

(2018) Chapter 6 – Airbnb and its competitors. In: Dolnicar

(ed), Peer-to-Peer Accommodation Networks: Pushing the Boundaries. Oxford, NY: Goodfellow Publishers, pp. 63–76.

40.

Hartmans

(2017) Airbnb now has more listings worldwide than the top five hotel brands combined. Inside Business. Available at: https://www.businessinsider.com/airbnb-total-worldwide-listings-2017-8 (Accessed June 23, 2022).

41.

Horn

Merante

(2017) Is home sharing driving up rents? Evidence from Airbnb in Boston. Journal of Housing Economics 38: 14–24.

42.

Hunt

Morgan

(1995) The comparative advantage theory of competition. Journal of Marketing 59(2): 1–15.

43.

Insideairbnb (2019) Get the Data. Available at: http://insideairbnb.com/get-the-data.html (Accessed 20 February, 2021).

44.

Kwoka

(1981) Does the choice on concentration measure really matter? The Journal of Industrial Economics 29(4): 445–453.

45.

King

Roberts

(2015) How robust standard errors expose methodological problems they do not fix, and what to do about it. Political Analysis 23(2): 159–179.

46.

Lerner

(1934) The concept of monopoly and the measurement of monopoly. Review of Economic Studies 1: 157–175.

47.

Moreno

Zhang

(2019) Agent pricing in the sharing economy: evidence from Airbnb. In: Hu

(ed), Sharing Economy. Cham, Switzerland: Springer, pp. 485–503.

48.

Liang

Choi

Joppe

(2018) Understanding repurchase intention of Airbnb consumers: perceived authenticity, electronic word-of-mouth, and price sensitivity. Journal of Travel & Tourism Marketing 35(1): 1–17.

49.

Liang

Schuckert

Law

, et al. (2017) Be a ‘superhost’: the importance of badge systems for peer-to-peer rental accommodations. Tourism Management 60: 454–465.

50.

Liu

Parkpian

(2018) Inbound tourism in Thailand: market form and scale differentiation in ASEAN source countries. Tourism Management 64: 22–36.

51.

Lorde

Jacob

Weekes

(2019) Price-setting behavior in a tourism sharing economy accommodation market: a hedonic price analysis of AirBnB hosts in the Caribbean. Tourism Management Perspectives 30: 251–261.

52.

Magno

Cassia

Ugolini

(2018) Accommodation prices on Airbnb: effects of host experience and market demand. The TQM Journal 30(5): 608–620.

53.

Majewska

(2015) Inter-regional agglomeration effects in tourism in Poland. Tourism Geographies 17(3): 408–436.

54.

Mody

Hanks

Dogru

(2019) Parallel pathways to brand loyalty: mapping the consequences of authentic consumption experiences for hotels and Airbnb. Tourism Management 74: 65–80.

55.

Möhlmann

(2015) Collaborative consumption: determinants of satisfaction and the likelihood of using a sharing economy option again. Journal of Consumer Behaviour 14(3): 193–207.

56.

Moulton

(1990) An illustration of a pitfall in estimating the effects of aggregate variables on micro units. The Review of Economics and Statistics 72(2): 334–338.

57.

Nieuwland

Van Melik

(2020) Regulating Airbnb: how cities deal with perceived negative externalities of short-term rentals. Current Issues in Tourism 23(7): 811–825.

58.

ONS (2019) Regional Economic Activity by Gross Domestic Product, UK: 1998 to 2018. Available at: https://www.ons.gov.uk/economy/grossdomesticproductgdp/bulletins/regionaleconomicactivitybygrossdomesticproductuk/1998to2018 (Accessed June 23, 2022).

59.

Pan

(2005) Market structure and profitability in the international tourist hotel industry. Tourism Management 26(6): 845–850.

60.

Peltzman

(2014) Industrial concentration under the rule of reason. The Journal of Law and Economics 57(S3): S101–S120.

61.

Porter

(1979) How Competitive Forces Shape Strategy. Harvard Business Review. Available at: https://hbr.org/1979/03/how-competitive-forces-shape-strategy (Accessed June 23, 2022).

62.

Prayag

Ozanne

(2018) A systematic review of peer-to-peer (P2P) accommodation sharing research from 2010 to 2016: progress and prospects from the multi-level perspective. Journal of Hospitality Marketing & Management 27(6): 649–678.

63.

Roberts

(2014) When bigger is better: a critique of the Herfindahl-Hirschman index’s use to evaluate mergers in networks industries. Pace Law Review 34(2): 894–946.

64.

Sainaghi

(2020a) The current state of academic research into peer-to-peer accommodation platforms. International Journal of Hospitality Management 89: 102555. DOI: 10.1016/j.ijhm.2020.102555

65.

Sainaghi

(2020b) Determinants of price and revenue for peer-to-peer hosts. the state of the art. International Journal of Contemporary Hospitality Management 33(2): 557–586.

66.

Saló

Garriga

(2011) The second-home rental market: a hedonic analysis of the effect of different characteristics and a high-market-share intermediary on price. Tourism Economics 17(5): 1017–1033.

67.

Schmalensee

(1977) Using the H-index of concentration with published data. The Review of Economics and Statistics 59(2): 186–193.

68.

Sen

(2005) Does increasing the market share of smaller firms result in lower prices? Empirical evidence from the Canadian retail gasoline industry. Review of Industrial Organization 26(4): 371–389.

69.

KKF

Min

(2018) Motivations and constraints of Airbnb consumers: findings from a mixed-methods approach. Tourism Management 67: 224–236.

(2017) Price determinants on Airbnb: how reputation pays off in the sharing economy. Journal of Self-Governance & Management 5(4): 53–80.

71.

Transparent (2020) Airbnb at IPO part III: how meaningful are Airbnb’s competitors? Available at: https://seetransparent.com/blog/airbnb-at-ipo-part-iii-how-meaningful-are-airbnb-competitors/ (Accessed June 23, 2022).

72.

Tussyadiah

(2016) Factors of satisfaction and intention to use peer-to-peer accommodation. International Journal of Hospitality Management 55: 70–80.

73.

Tussyadiah

Pesonen

(2016) Impacts of peer-to-peer accommodation use on travel patterns. Journal of Travel Research 55(8): 1022–1040.

74.

Wang

Nicolau

(2017) Price determinants of sharing economy based accommodation rental: a study of listings from 33 cities on Airbnb.com. International Journal of Hospitality Management 62: 120–131.

75.

Xie

Mao

(2017) The impacts of quality and quantity attributes of Airbnb hosts on listing performance. International Journal of Contemporary Hospitality Management 29(9): 2240–2260.

76.

Xie

Mao

(2019) Locational strategy of professional hosts: effect on perceived quality and revenue performance of Airbnb listings. Journal of Hospitality & Tourism Research 43(6): 919–929.

77.

Zervas

Proserpio

Byers

(2017) The rise of the sharing economy: estimating the impact of Airbnb on the hotel industry. Journal of Marketing Research 54(5): 687–705.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.27 MB