Innovation in tourism

From innovation to tourism innovation,…and back to innovation

Nearly all the studies on tourism innovation start by recognizing that, while an inevitable point of departure, the classical Schumpeter’s (1955) definition of innovation is mainly technological in nature and thus too narrow to embrace its occurrence in tourism. A wider conceptualization is retained necessary to account for those non-technological innovations, which pervade tourism as a service sector, in particular in the form of new organizational and marketing practices (Camisón and Monfort-Mir, 2012). Accordingly, the ‘dual-core’ model proposed by Daft (1978) – distinguishing between innovations in the ‘technological core’ (‘technical system’) and in the ‘administrative core’ (‘social system’) of the firm – and its operative translation into the third edition of the Oslo Manual (OECD, 2005) – for the measurement of scientific and technological activities – is often taken as a more reliable conceptual reference to deal with tourism innovation (Hjalager, 2010). On the other hand, much of the extant literature feels uncomfortable also with this latter ‘service-friendly’ definition of innovation, in terms of ‘implementation of a new or significantly improved product (good or service), or process, a new marketing method, or a new organizational method in business practices, workplace organization or external relationship’ (OECD, 2005). While wider, this conceptualization and the measurement methodologies that are based on it – for example, the Community Innovation Survey (CIS) in Europe and several composite innovation indicators at the international level (by the World Economic Forum, World Bank and the like) – would not be capable to capture more granular forms of change, which have been claimed to represent ‘innovations’ other than ‘pure’ marketing and organizational ones in tourism (Camisón and Monfort-Mir, 2012), like changes in: the modalities tourism services are provided (Ottenbacher and Gnoth, 2005) and commercialized (Gronroos, 1990); the relationships through which tourism companies interact with their customers and with other actors of the value chain; the systems of institutions tourism players are embedded in, especially at the territorial (e.g. tourism destination) level (Hall, 2009); and the business models tourism companies adopt to exploit their innovations (Hjalager, 2002).

The way in which tourism research has tried to ‘crystallise’ this last kind of observed changes into innovations is twofold. On the one hand, more encompassing definitions of innovations have been provided, which extend both the ‘form’ (i.e. the functional domain in which they manifest) and the ‘impact’ (i.e. the geographical domain) of the changes, which could be considered as innovative in tourism (on this, see e.g. the wide definitions provided in Hall and Williams, 2008). On the other hand, new ways of defining the kind of change that would amount to a tourism innovation have been proposed. For example, Camisón and Monfort-Mir (2012) have suggested that innovation in tourism should refer to the ‘capacity’ to change – that is, the ‘dynamic capabilities’ through which firms learn to face change (Eisenhardt and Martin, 2000) – rather than to the outcome/performance of the same capacity, to which the Schumpeterian approach instead alludes. Along a similar vein, Nordli (2017) proposes to look for tourism innovation in the process of value co-creation – with suppliers and customers – that typically constitutes the ‘service logic’ (Grönroos and Voima, 2013). Keeping the focus on innovation as output/performance, a new conceptualization of it has been proposed by looking at tourism through the lens of the ‘experience economy’ (Pine and Gilmore, 1999). According to Volo (2006), those technological and non-technological changes that the providers of a ‘tourism experience’ introduce in any component of the ‘tourism bundle’ can be deemed innovation providing they affect the experience of tourists in one of its constitutive dimensions (‘accessibility’, ‘affection’, ‘convenience’ and ‘value’). Along the same experience-based approach to tourism (Stamboulis and Skayannis, 2003), the ideas of ‘experience innovation’ and of ‘experience concept innovation’ have been put forward and investigated in specific tourism segments (e.g. Eide et al., 2017; Eriksen, 2015; Jernsand et al., 2015).

Both of the above approaches represent interesting attempts to identify tourism-specific forms of innovation, which would remain otherwise ‘hidden’ (Camisón and Monfort-Mir, 2012; Nordli, 2017). On the other hand, both of them have been implemented by ‘diverging’, not only from a narrow Schumpeterian definition of innovation, which the extant literature on innovation has also progressively extended (e.g. through the different versions of the Oslo Manual). But divergence has also occurred from the Schumpeterian conditions that make a ‘simple’ kind of change an actual innovation, which the innovation literature instead keeps retaining pivotal. The first of these conditions is the novelty and the reproducibility of an innovative change, which the extension (in Drejer’s words, ‘the infection’) of the forms of innovation to be included in tourism (as well as in services) has been made hard to inspect. As Hjalager (2010) noticed in her first and seminal review of the literature,

[i]n practice and in many of the empirical studies, enquirers are willing to accept that innovation is generally characterised by everything that differs from business as usual or which represents a discontinuance of previous practice in some sense for the innovating firm (p. 2).

Even more divergent is the choice about the relevance that the Schumpeterian theory attributes to the capacity of producing a viable economic impact, for a change to be considered an actual innovation. On the one hand, the scanty number of systematic studies that Hjalager (2010) had detected about the economic impact of innovation in tourism enterprises and destinations has remained so also in the following years (Omerzel, 2016), and is still dominated by non-generalizable case-based evidence (see ‘The empirical analysis of tourism innovation and its ‘biases’ section below). On the other hand, the new conceptualizations of tourism innovation, in terms of capabilities and experience, have made the search for an economic impact of secondary importance in the identification of tourism innovation, and focused the attention on a less strictly economic kind of effects.

On the basis of the previous considerations, it seems plausible to argue that, in looking for a more tourism-sensitive definition/concept of innovation, tourism research has actually ended up by diverging more substantially from the economic analysis of innovation in some key conceptual requisites, which should however invariably apply also to its more specific declinations.

Classifying and reclassifying tourism innovation

The opportunity of departing from the Schumpeterian tradition has also been claimed by tourism studies in front of the typologies of innovation that its original and extended definitions (see ‘From innovation to tourism innovation,…and back to innovation’ section) have led to identify. First of all, the standard four categories of the Oslo Manual (in its latest edition) – that is, product, process, organizational and marketing innovations – have been re-elaborated in order to adapt to the specificities of tourism (Hjalager, 2010). The most substantial re-elaboration has concerned the category of process innovations, which has been basically ‘reduced’ to the adoption of externally developed information and communication technologies (ICTs) in the different subsectors of the industry, lodging (e.g. hotels and restaurants) and transports (e.g. airports) in particular (Rodgers, 2007; Sheller and Urry, 2006). More in general, given the experience-based nature of most product tourism innovations – one might think of the low-cost ‘formulas’, recently introduced in the hotel sector, by reconfiguring the balance between the main tourists’ experience dimensions (‘accessibility’, ‘affection’, ‘convenience’ and ‘value’) – process innovations have become the equivalent of technological innovations, which would be hardly identifiable in other spheres of the innovation process (Omerzel, 2016). A certain readaptation has also concerned the two standard categories of non-technological innovations of the Oslo Manual, whose functional boundaries have been either reduced – organizational innovations in tourism would be mainly limited to ‘managerial innovations’, in the way the workforce is administered and motivated (Hjalager, 2010) – or expanded – in the tourism sector, marketing innovations are claimed to be part of wider ‘management innovations’, which would include changes in the organizational policies and culture of the providers (Vila et al., 2012).

A second methodological choice of tourism research has been that of adding to the four standard categories of (manufacturing) innovation some additional ones, which could capture its unfolding in alternative domains (Weiermair, 2006). A first addition has occurred in parallel to that recently experienced by the literature on manufacturing and service innovations, and concerns the concept and modes of eco-innovation (Marzucchi and Montresor, 2017), which however encompasses, rather than alternating, the four original categories (in particular, product and process innovations). The increasing importance of environmental sustainability has actually stimulated the interest for those innovations through which the environmental impact of tourism firms and destinations can be alleviated (e.g. Hall, 2008; Bell and Ruhanen, 2016; Liu et al., 2017; Martínez-Pérez et al., 2015). A more specific addition is that of ‘institutional innovations’ (Hjalager, 2010: 3). In their variegated manifestations (e.g. Van Wijk et al., 2015), these are intended to capture a wide category of changes occurring in both the collaborative/organizational structures through which tourism players interact (e.g. their networking) – an aspect to which the standard definition of organizational innovation already refers to, in terms of ‘external relationships’ – and in the legal and regulatory framework within which they operate (e.g. franchising and licensing arrangements) – referring to cases of institutional change that however have wider implications than on the tourism sector.

The search for alternative and/or new classifications, with a superior interpretative power than the existing ones, does not represent per se a substantial divergence with respect to the broader analysis of innovation in manufacturing and services, where a similar effort has been made since long. One just needs to recall the notable distinctions between ‘architectural’ and ‘modular’ innovations in the technology/manufacturing domain (Henderson and Clark, 1990), or that between ‘competence destroying’ and ‘competence enhancing’ ones (Anderson and Tushman, 1990). What is more diverging is the way the work of reclassification has been done in tourism research. First of all, unlike in ‘mainstream’ innovation studies, the re-elaborated versions of the existing categories as well as the new ones have not been crystallized yet into distinguishing definitions, which could help to draw the boundaries, at least in principle and conceptual terms, between themselves – let us think of the partial overlapping between managerial and management innovations – and between them and standard innovation categories – the substantial overlapping between the ‘new’ institutional and the ‘old’ organizational innovations is a case in point. As innovation studies have shown, such a definitional effort is either prodromal to the attempt at measuring specific innovations – let us think of the relationship between the Schumpeter theory of innovation and the Oslo Manual – or posthumous to their empirical detection and investigation – like in the case of the latest typologies of technological innovations we have recalled above. One way or the other, defining innovation categories/typologies is essential to investigate their actual reproducibility and diffusion and, what is more, their actual relevance. This is an additional element of divergence of the research on tourism innovation, which at least so far has mainly used the new proposed categories as a sort of ‘shopping-list’ of innovation in tourism, without engaging in the systematic search of their diffusion and of their eventual distinguishing impact. In innovation studies, this is actually the main raison d’être of different innovation types, whose differential impact has been searched and found in several respects – let us think of the differential impact of product and process innovation on productivity (Hall, 2011) and on job creation/destruction (Bogliacino and Pianta, 2010).

A second divergent choice in terms of innovation typologies in tourism refers to their mutual relationships. In the majority of the studies that engage in their proposal and/or analysis, the point emerges soon about their pervasive co-occurrence. Like in other services, and possibly to a greater extent, in the tourism sector different categories of innovation occur simultaneously and interplay between them (Weiermair, 2006). Just to make an example, process innovations (ICT adoption) and marketing innovations have been shown to interact cumulatively in the airlines sector (Buhalis, 2004). Their interplay is pervasive also in other sectors and ‘certainly lead[s] to the breakdown of borders between categories’ (Hjalager, 2010: 4), making them possibly more ‘fuzzy’ than in services in general (Gallouj and Weinstein, 1997). This is actually the main implication of innovation typologies to which tourism research has paid attention so far. In the experience-based literature, for example, both consumer-based tourism innovations (Volo, 2006) and experience-concept innovations (Eide and Mossberg, 2013) have been characterized as bundles of innovation typologies (product, process, organizational and market), whose individual role and mutual relationships are retained unrecognizable. In the measurement of tourism innovation, even the most basic innovation typologies (i.e. product vs. process innovations) have emerged hard to be disentangled by survey respondents, and the proposal has been accordingly made to avoid their use (Nordli, 2017). This concern is for sure motivated. On the other hand, its exclusive consideration represents a further step away from the innovation literature, which has also paid attention to the fluid boundaries between innovation typologies since long (on that between product and process innovation, see e.g. Simonetti et al., 1995). While recognizing the importance of the issue, innovation studies have however also tried to investigate the extent to, and the modalities through which, different categories of innovation interact between them, especially technological with non-technological ones. The streams of research on the so-called ‘complex innovators’ (Evangelista and Vezzani, 2010) and on the complementarities across innovation types (Carboni and Russu, 2018) represent two notable instances of such an effort, which has found only limited correspondence in tourism (and service) innovation so far (see, e.g. Guisado-González et al., 2014).

As it emerged in dealing with the conceptualization of innovation (see ‘From innovation to tourism innovation,…and back to innovation’ section), also with respect to its classification, tourism research seems to have gone beyond a justifiable effort of differentiation. Not only are the identified typologies different and/or rearranged with respect to the standard ones of innovation studies. But their identification has been pursued by diverging from them in some important methodological respects – that is, boundaries definition, relevance and impact and mutual relationships – which would be arguably possible and opportune to converge with.

Determinants and effects of tourism innovation: Taking innovation theories seriously

In parallel with the research progress on the conceptualization/classification of tourism innovations, an appreciable effort has also been put in the search of their determinants and in the analysis of their effects. The attention to these issues detected by Hjalager’s (2010) review has continued over time (Omerzel, 2016) and is visible also in the latest studies on the topic (e.g. Grissemann et al., 2013; Naidu et al., 2014). A certain continuity can also be found in the theoretical foundations that have been driving such an analysis, which have remained weakly developed and, using Hjalager’s (2010: 4) words, mainly ‘implicit’. As we will argue in the following, this represents a further element of divergence from innovation studies, which is in turn due to an only partial kind of reference to their theoretical contents.

At the outset, referring to the innovation literature for the theories at stake appears for tourism scholars nearly inevitable. In spite of the substantial work undertaken – especially in the analysis of experienced-based kinds of innovation – tourism innovation still lacks of a coherent body of autonomous theoretical insights, which could account for its specific drivers and impacts with respect to manufacturing and service innovation in general. A theoretical foundation similar to that through which the economics of service has been able to differentiate its innovation analysis from that of manufacturing (Gallouj and Savona, 2009) is actually still missing. On the other hand, the reference of tourism research to ‘standard’ innovation theories has been mainly functional to the ‘simple’ identification of candidate explanatory factors – for example, the driving role of ICT, as from the ‘technology-push’ approach (e.g. Naidu et al., 2014) – and plausible impact domains of tourism innovation – for example, the competitiveness impact of tourism, as from the resource competence-based view of the firm (e.g. Grissemann et al., 2013). The deeper implications of these theories for the mechanisms through which tourism innovations are driven and exert their economic impact have been instead only superficially addressed.

As far as the determinants are concerned, their analysis has been carried out by drawing eclectically on different theoretical domains, which represent quite consolidated streams of innovation literature, that is (i) innovative entrepreneurship, (ii) the technology push/demand pull debate, (iii) organizational (competence and resource-based) knowledge, (iv) research-based knowledge, (v) localized knowledge and (vi) innovation systems (this is the picture reconstructed by Hjalager, 2010). However, in the majority of the cases, the link has been limited to the front topic of each stream, without searching for more specific connections. Some examples could help to detect this point, at a progressively more aggregated level of analysis. To start with, the pivotal role that the ‘lifestyle entrepreneur’ plays in tourism innovation (Ateljevic and Doorne, 2000) could be more extensively disentangled by linking its analysis to the literature on the relationship between entrepreneurial features (e.g. age, gender, family status, etc.) and innovation (Nightingale and Coad, 2014). As for organizational knowledge, the increasing attention that tourism research has been paid to the firm’s absorptive capacity and to external knowledge acquisition (e.g. Martínez-Pérez et al., 2016; Thomas and Wood, 2014) could be enriched by referring to the latest developments about the manifold nature of the same capacity (e.g. Franco et al., 2014) and, above all, about the so-called ‘open innovation mode’ (Chesbrough, 2006). As for localized knowledge and innovation systems (e.g. Carson et al., 2014), useful insights about the unfolding of tourism innovation at specific destinations could be obtained by bridging its analysis with that of innovation in economic geography, particularly in its evolutionary declination (Boschma and Frenken, 2006).

The reference to innovation studies has been partial also in the investigation of the impact of tourism innovation, which has been explored in different domains (e.g. financial performance, quality standards, market sales, just to mention a few (Omerzel, 2016)), but with little theoretical justification. This is for sure the case of its productivity impact (e.g. Blake et al., 2006; Torrent-Sellens et al., 2016), so far addressed without encompassing the articulation of direct and indirect effects that have been instead recognized with respect to manufacturing innovation, such as in the famous CDM model by Crépon et al. (1998) (for a recent attempt to set tourism innovation in this framework, see Divisekera and Nguyen (2018)). The same holds true for the impact of tourism innovation on the market value of firms (e.g. Nicolau and Santa-Maria, 2013), which would benefit from a more integrated consideration of the analysis that innovation studies have made of how financial markets value knowledge-intensive assets (Hall, 1999).

The limited extent to which the extant research has looked for the theoretical foundations of the antecedents and consequences of tourism innovation is connected to another diverging aspect: the differential attention innovation policy has found in tourism, as mainly prescriptive and with only ‘vague evidence of its effects and effectiveness’ (Hjalager, 2010: 8). If we exclude networking and collaboration policies – which the small-medium-enterprises (SMEs) typical of the sector have naturally induced to inspect (Novelli et al., 2006) – a mapping of the market and non-market failures that require one rather than another policy intervention is in fact still missing (Halkier, 2014; Rodríguez et al., 2014). As a consequence of that, the literature still misses a conceptual framework on the basis of which to look for the `additionality' of tourism innovation policies. Their evaluation is thus another important gap to be filled (Hall and Williams, 2008).

In conclusion, while tourism research has drawn on innovation studies in dealing with the determinants and the effects of tourism innovation, a divergence remains in the way their analysis has been grounded into sound theoretical foundations. The choice of making innovation theories mainly functional to the identification of the candidate drivers and effects of tourism innovation, and to refrain from developing more articulated hypotheses about their specific functioning, represents another element of divergence, which would still be opportune to overcome.

The empirical analysis of tourism innovation and its ‘biases’

The most apparent terrain of divergence in the analysis of tourism innovation is for sure represented by the empirical methodologies through which evidence about them is collected and investigated, especially about their determinants and economic impact. A first deviation concerns the balance between qualitative and quantitative research. Although a dedicated survey of the adopted empirical methodologies would be necessary and welcomed to accurately ascertain it, the existing reviews of the literature and the latest contributions reveal that the relative weight of qualitative methods (e.g. case studies, semi-structured in-depth interviews, ethnographic studies and the like) over quantitative ones (i.e. statistical and econometric analyses based on primary or secondary data) is substantially larger in tourism than in innovation studies. In the first decade of tourism innovation research, covered by Hjalager’s (2010) review, the reference to piecemeal case studies has been possibly motivated by the need of identifying, inductively as well as deductively, the distinguishing features of a still unexplored phenomenon. For similar reasons, the use of in-depth qualitative methodologies has also been crucial for the take-off of innovation studies in general (Martin, 2016). On the other hand, while consolidating its theoretical premises, the innovation literature has also progressively moved towards a systematic quantification of the innovation processes. In so doing, it has equipped with specific measurement methodologies and analytical tools to ascertain their actual consistence, determinants and impacts, and to draw reliable predictions as well as policy/strategic recommendations. Conversely, while detectable to a certain extent, the effort of tourism research in the same direction has been less intense. According to the most recent review of the literature by Omerzel (2016), the relative incidence of quantitative over qualitative studies appears to have grown over time (more than 45% of the 182 papers of his survey are characterized as quantitative). Still, the search for a general and generalizable kind of evidence about tourism innovation appears less a priority than in innovation studies, which do not appear a benchmark in this last respect either. Such an element of divergence is partially explained by the lack of suitable ‘secondary’ data (e.g. by international organizations, like the EC, the UN and the World Bank) for a systematic measurement and investigation of tourism innovation. Indeed, their collection is still inspired by theoretical approaches that mainly fit with the Schumpeterian paradigm of manufacturing innovation and leave much of tourism (and service) innovation ‘hidden’ (Camisón and Monfort-Mir, 2012). However, an important part of the deviation at stake does not seem driven by data constraints and apparently responds to a choice of camp. First of all, in spite of their bias/focus, existing international datasets on innovation do already contain information on tourism firms/sectors (e.g. some releases of the CIS and of the EC Innobarometer), which could be exploited more intensively at least to investigate the ‘non-hidden’ part of their innovations (i.e. of technological, organizational and marketing nature), and the same holds true for some country-specific business data sets (e.g. the Business Longitudinal Database developed by the Australian Bureau of Statistics). Second, the search for alternative measurement frameworks, which could enable to bring this hidden innovation in surface, is still limited to some ‘local’ proposals (e.g. Nordli, 2017, with respect to Norway) and hesitates to involve international organizations, which could take care of their codification into manuals and of their utilization. Third, the collection of primary data on tourism innovation, in which several scholars have engaged to remedy the limitations of secondary ones – another aspect in need of a dedicated review– is often designed in such a way to reach an only limited level of sampling coverage and representativeness of the reference population, thus hampering the achievement of generalizable results.

The last consideration is connected to a second element of divergence in the use of empirical methodologies, which refers to the ‘sophistication’ of the analytical tools adopted in quantitative research on tourism innovation. Once again, a survey about these methodologies would be desirable to support more strongly what emerges from a general appreciation of the econometric analysis of tourism innovation. Econometric strategies are generally aimed at detecting correlations among focal variables, rather than proper causality relationships between them, either in terms of antecedents or effects, on which innovation studies are instead focusing increasingly more. In this case too, the lack of proper data is partially responsible for the divergence at stake. The implications that a more intensive use of primary versus secondary sources has for the limited availability of longitudinal data are for sure one of the most remarkable motivations (an interesting exception to this pattern is the recent longitudinal study by Divisekera and Nguyen (2018)). On the other hand, another choice of camp with respect to innovation studies seems to emerge from the still limited resort of the research on tourism innovation to the analytical instruments through which the ‘spurious’ nature of the investigated relationships – again, of both innovation causes and effects – is addressed. Indeed, a number of these tools are amenable to be used also in a cross-sectional environment (one just needs to think of variable instrumentation or selection and simultaneous equations), and could definitively help in getting closer to more solid empirical relationships.

All in all, while the available measurement methodologies and the kind of data collected through them provide an important justification to the differentiation of the empirical research between general innovation and tourism innovation, part of this differentiation seems to result from another set of diverging choices, which add to the previous ones and to whose implications we will turn in the next conclusive section.