The seven lives of Airbnb. The role of accommodation types

Highlights

• Travellers are very reluctant to book shared flats on Airbnb during the pandemic.

• Full flats in Airbnb – controlling for price levels - are preferred to hotel rooms.

• Need for physical distance explains the enhanced propensity to book full flats.

• Being able to change physical distance perceptions rebalances tourists' choice.

• We provide actionable suggestions on how to maximise bookings despite the pandemic.

Abstract

This paper investigates how COVID-19 is impacting different accommodation types, and whether travellers' choices regarding accommodation type are affected by the need for physical distance. Study 1 shows that travellers are very reluctant to book shared flats on Airbnb during the pandemic. However, full flats – controlling for price levels - are preferred to hotel rooms. Study 2 clarifies the process behind the increased choice for full flats, i.e., the need for physical distance. In Study 3, we actively manipulate physical distance and show that assuring physical distance will reduce the concerns towards hotel and shared flat options. Apart from enlightening the psychological process behind accommodation choice, the study offers operators actionable suggestions on how to maximise bookings despite the pandemic.

Introduction

The global pandemic has posed a serious threat to the global economy, the tourism and hospitality sector in general (Sharma & Nicolau, 2020), and to paid peer-to-peer accommodation types specifically (Dolnicar & Zare, 2020). Peer-to-peer models have grown exponentially in recent years (Ert et al., 2016; Gutiérrez et al., 2017; Viglia, 2020), mainly due to rapid technological developments (see Buhalis et al., 2019). Over time, paid peer-to-peer models like Airbnb have threatened traditional hotels by stealing market shares, proposing new ways to access services and focussing on unique experience, social/physical interactions and lower prices (Tussyadiah and Pesonen, 2016, Tussyadiah and Pesonen, 2018). In line with this, a growing body of literature has recently discussed the differences between paid peer-to-peer versus traditional models (i.e., Heo, 2016; Osman et al., 2019), along with the elements that travellers look for when choosing one model over another (Hajibaba & Dolnicar, 2017). Additionally, some systematic reviews on the state of current research (see Dolnicar, 2019; Prayag & Ozanne, 2018; Sainaghi, 2020; Sainaghi & Baggio, 2020) show substitution effects between these two realms.

Traditional models (e.g., hotels) are characterized by strong institutional settings (Osman et al., 2019), while in the case of paid peer-to-peer accommodation there is a lower social/physical distance and higher trust between guests and hosts (Ert & Fleischer, 2019; Pera et al., 2019). Hence, traditionally, travellers who choose a paid peer-to-peer model seek meaningful social/physical interactions with locals, unique experiences and lower prices (Tussyadiah & Pesonen, 2016). As argued by Dolnicar (2019, pp. 248–249), ‘a unique feature of paid online peer-to-peer accommodation is the wide range of different kinds of spaces being offered’ and ‘sometimes the human relationship rather than a house is… the primary shared asset’. To the best of our knowledge, only one study has explored empirically different types of paid peer-to-peer options, such as shared rooms, private rooms and entire homes (Tussyadiah, 2016). In particular, the author found that the pursuit of social benefits is associated more with the choice of shared flat than full flat. Nevertheless, as anticipated, the COVID-19 pandemic has changed the status quo of the sector, imposing new dynamics regarding the structures of accommodations, on the one hand, and the choices, needs, necessities and uses of travellers, on the other (Gössling et al., 2020). Consistently, Dolnicar and Zare (2020) have recently proposed that COVID-19 will disrupt the paid peer-to-peer accommodation type, and recent data show that paid peer-to-peer services have witnessed a dramatic drop in business volumes during the lockdown (Glusac, 2020). Moreover, given the economic crisis and the continued social/physical distancing applied in many countries around the world, these business models will likely continue to suffer for many months. Brian Chesky, Airbnb's chief executive, declared: ‘Airbnb's business has been hit hard, with revenue this year forecasted to be less than half of what we earned in 2019’. Nonetheless, Airbnb will now be listed in the stock market this year, which shows that its future is far from over (Driebusch et al., 2020).

In this paper, we propose and hypothesize that the paid peer-to-peer accommodation type is not dead but that there will be a change in bookings in favour of full flats rather than shared flats. In other terms, the pandemic reduces tourists' expectations and the search for the social/physical interaction typical of peer-to-peer models (Osman et al., 2019; Pera et al., 2019) and hotel main receptions and common spaces, such as bars or lounges. Along these lines, the pandemic favours types of accommodations that guarantee physical distance, such as full flats. Accordingly, we also hypothesize - and find evidence for it - that the increased choice for full flats is driven by the need for physical distance.

Physical distancing asks for fast changes in the social habits of citizens (Pfattheicher et al., 2020), so the physical aspect can be seen as the driving social habits change due to the pandemic. We use the composed term “social/physical” throughout the paper to reflect the physical part of social interactions (vs. social/technology-mediated interactions). In the past, social and physical interactions have been regarded mainly as synonymous/interchangeable, with the second being the main vehicle for the first (conviviality is a clear example). In the digital age (as exemplified by travellers' blogs or online communities), we feel it important to specify what we refer to.

The scientific community has found and communicated that COVID-19 can occur in specific settings, particularly indoor, where infected person(s) spend long periods of time with others (World Health Organization; https://www.who.int). A lingering question for the International health authorities and policy makers has therefore become “how to reduce virus spread while limiting also the negative socio-economic impact”. Physical distancing, that is minimizing one's contacts with others, has been proposed as one of the useful measures. Depending on the specific geographical context and on the timing, physical distancing has been induced through mandatory regulations (such as lockdown) and/or encouraged as voluntary practice by calling upon each citizen self-responsibility.

Measures such as physical distancing ask for fast changes in core social habits of citizens (Pfattheicher et al., 2020). Minimizing contacts between people may lead to benefits that are both egoistic in nature (namely, protecting oneself) and altruistic or social (as protecting others). At the same time, they entail personal costs. The presence of physical distance, indeed, inhibits also social/physical interaction. Given these arguments, we portray that need for physical distance is a key factor driving changes in each one's behaviour, including tourists' consumption patterns.

Methodologically, the paper proposes, in line with previous studies (Pera et al., 2019; Tassiello et al., 2018), three experiments to test the choice of the type of accommodation (full flat, shared flat, and hotel) across two different scenarios (non-pandemic versus pandemic one).

We test how these choices are mediated by the need for physical distance. We also actively manipulate physical distance to see whether a change in the level of this variable is able to influence accommodation choice. As far as we know, it is the first time that a tourism paper proposes a manipulation of a mediator in experimental designs (see Jacoby & Sassenberg, 2011). Using a manipulation of a mediator is important because it allows assessing whether it is possible to actively affect a process that induces certain behaviours (e.g., the choice of an accommodation). The empirical research finds that: a) controlling for price levels, full flats gained popularity during the pandemic crisis compared to hotels and shared flats; b) one person's need for physical distance represents the underlying mechanism behind the enhanced choice for full flats on Airbnb during the pandemic; c) the need for physical distance contributes to explaining the relationship between the pandemic situation and the accommodation type chosen, such that the ability of guaranteeing physical distance increases the preference for shared rooms and hotels.

The paper provides two clear implications for theory. First, the paper contributes to the literature on paid peer-to-peer accommodation types (e.g., Dolnicar, 2020; Dolnicar & Talebi, 2020; Ert & Fleischer, 2020; Guttentag et al., 2018), suggesting that this model is not dead but rather will undergo a major change. Specifically, the paper adds to the literature on social/physical distance in the accommodation realm (Osman et al., 2019; Pera et al., 2019), highlighting that, due to the COVID-19 pandemic, travellers will increasingly choose the full flat model. Second, drawing from spatial distance theory (Williams & Bargh, 2008), this paper has proposed – and found evidence for – the need for physical distance (Pfattheicher et al., 2020) as an important underlying mechanism. We expect that the need for physical distance will affect travellers and influence their decisions and behaviours in the coming months and years, due to the COVID-19 pandemic.

From a managerial perspective, the study encourages to: a) list entire homes instead of splitting them into different listings; and b) put emphasis on, and include in communication, all the measures and practices to increase physical distance between strangers such as differentiated pathways. Overall, the paper sheds light on the emotional and behavioural responses of tourists to the changes imposed by the Covid-19 pandemic.