The Decline of Tourist Destinations: An Evolutionary Perspective on Overtourism

: The term overtourism has generated considerable attention both in academic discourse and public debate. The actual or perceived impact of overtourism on destinations has signiﬁcant ecological, social, and cultural consequences. However, a crucial question remains unanswered: What does overtourism do to a destination’s tourism industry itself? At the core of this question is whether overtourism is a self-limiting phenomenon or a cumulative one, and how precisely overtourism shapes patterns of quantitative or qualitative decline of a destination’s tourism sector. This article o ﬀ ers a conceptual discussion of the impact of overtourism on a destination’s local tourism sector by reﬁning the latter stages of Butler’s tourist area lifecycle through forms of path decline known from evolutionary economic geography. By combining these two theorical approaches and reﬁning the typology of path decline from evolutionary economic geography to the case of tourism under an overtourism scenario, this article suggests that, in the absence of exogenous changes due to policy interventions or public pressure, under an overtourism scenario, a destination’s tourism sector might contract, downgrade, dislocate, and eventually even disappear. Further research should focus on how to prevent these forms of path decline.

As such, the overtourism debate has come to revolve around contradictions between tourists' and inhabitants' needs and interests and resulting questions of social and ecological sustainability [8], drawing on the long-standing debate about how forms of sustainable or responsible tourism might look [9,[21][22][23][24]. The question of how precisely these negative phenomena, understood under the umbrella term "overtourism", affect the tourism sector in a destination has rarely been asked. If, for example, overtourism basically means an interior sustainability deficit within the tourism sector and leads to a situation of "tourism killing tourism" [9] (p. 157) or destroying its own foundations [4] (p. 399), the result would be a deterministic lifecycle pattern that results in a self-limiting course of overtourism. However, this does not seem to be the case, as the strong growth of tourism in Barcelona, probably the most widely discussed case of Mediterranean overtourism, demonstrates [7] (p. 8). Instead of overtourism limiting itself through numbers of visitors in affected destinations returning to a sustainable level, there seems to be a need for policymakers to actively constrain excessive growth, as is evident in Dubrovnik's 8000 visitor limit [4] (p. 399). Could it be that in the absence of active policy interventions, instead of being a self-limiting phenomenon, the problems related to overtourism actually exhibit a cumulative nature? Such a cumulative nature of overtourism would basically mean a lock-in pattern of the tourism sector in a suboptimal development trajectory, a phenomenon discussed in economic geography, mainly in relation to manufacturing sectors [25], but in some instances also for tourism [26,27].
Indeed, despite the lively public and academic debate on overtourism, surprisingly little is understood about processes of evolution within the tourism sector in affected destinations. Butler's [28] well-known model of the tourist area lifecycle is still useful as a concept of how tourist destinations may evolve, particularly since it offers different trajectories at the end of the lifecycle. However, the tourist area lifecycle remains rather vague in terms of the trajectories that may ensue at the cycle's end, thus calling for refinement from an evolutionary perspective [26] (p. 95). Furthermore, it is not yet understood how overtourism or, more precisely, excessive, unsustainable and/or undesirable forms of tourism affect the course of the evolution of a destination's tourism sector, notwithstanding the wider aspects of social and environmental sustainability. While the latter aspects are undoubtedly important and, in reality, cannot be isolated from the evolution of the tourism sector in a narrow sense, to advance our understanding of how "overtourism" affects the evolution of tourist destinations, it makes sense to isolate the evolutionary trends that could be at play within the tourism sector of affected destinations. This article attempts to do so by further refining the trajectories at the end of Butler's tourist area lifecycle [28] and by drawing on forms of path decline recently identified in evolutionary economic geography [29], focusing these trajectories on problems prevalent in an overtourism scenario. Indeed, the usefulness of applying concepts of evolutionary economic geography to tourism has been demonstrated before [26,27]. The article's aim is therefore to identify stylized trajectories that link phenomena related to an overtourism scenario with the decline of a destination's local tourism sector. The article pursues a conceptual approach that is meant to provide an impetus for in-depth empirical research into how precisely the pathways of a destination's tourism sector unfold under an overtourism scenario and on how to shift towards a more beneficial pathway. While, due to the conceptual approach followed, the article can provide only an impetus for further research, avenues for further research and, notably, for empirical research in destinations at risk of path decline are explored.
The article is structured as follows. The next section presents a brief review of the overtourism literature, with the aim of developing a sufficiently clear working definition of overtourism to guide the arguments presented and to distinguish overtourism from other concepts such as mass tourism. Then, the article develops an evolutionary perspective on tourism development by reviewing both the tourist area lifecycle model and recent approaches in evolutionary economic geography and specifically in the literature on regional industrial path development. The article goes on to discuss a stylized typology of path decline in the tourism sector of destinations affected by an overtourism scenario. Finally, the paper lays out some conclusions for tourism policy and destination management in affected destinations to prevent path decline, and sketches avenues for further research.

Overtourism: A Brief Conceptual Review
As Koens et al. [2] highlight, the debate about "overtourism" addresses a number of multidimensional phenomena, not all of them strictly limited to tourism itself [6] (p. 5), and suffers from a lack of analytical clarity [2]. Furthermore, the debate is not free from myths and popular perceptions due to the high attention afforded to these phenomena by the popular media [2,6]. Still, what the "overtourism" debate generally revolves around can indeed be grasped, as Koens et al. do when they sum up the matter as "an excessive negative impact of tourism on the host communities and/or natural environment" [2] (p. 2). The World Tourism Organization understands overtourism as phenomena that exert an excessive negative impact on citizens' subjective evaluation of the local quality of life or on the quality of the tourism experience [6] (p. 4).
Despite the lack of clarity in the term "overtourism", which, to some degree, makes it a "fuzzy concept" [30] due to its imprecise definition [2] (p. 2), it is still possible to capture some of the phenomena commonly associated with the term and thus to define what typically characterizes a scenario in which a destination might find itself suffering from some form of "overtourism". Following, in particular, Koens et al. [2] (p. 5), among these phenomena are overcrowded public areas in affected destinations, the undesirable behavior of tourists causing noise or other annoyances such as the highly visible "beer bikes" [2] (p. 6), the touristification of stores and amenities, pressure on the housing market in view of the renting out of residential apartments via sharing-economy online accommodation platforms, the decreasing relative purchasing power of inhabitants, and environmental damage such pollution or increased strain on waste collection and water supply [1] (p. 4) [2] (p. 5) [3] (p. 554) [4] (pp. 399-401) [5] (p. 375) [11,12,19,31].
As these phenomena demonstrate, "overtourism" is not just about how much tourism occurs in a destination (e.g., how many tourists arrive at a destination or stay there for a number of days), but is perhaps even more about precisely what kind of tourism occurs, how it does, and how it is being managed [6] (p. 5). This is why overtourism, while related to masses of tourists and their growth [10] (p. 353), is different from the concept of mass tourism [2] (p. 9). Mass tourism, too, is a complex concept that includes a number of different images such as the Fordist model [32,33] of highly standardized package tourism, but also more differentiated forms of mass tourism [34]. Spatially, Mediterranean mass tourism in its package variant is often reflected in small-scale tourism agglomerations [35] (p. 1) [36] (p. 39) that do not lend themselves to the kind of frictions between visitors and residents that are usually at the focus of the overtourism debate-that is, frictions caused by larger number of tourists intruding into semi-private spaces of local residents or into public urban spaces in residential neighborhoods [1] (p. 3) [2] (pp. 6-7) [8]. In consequence, overtourism is not a problem of mass tourism per se but of particular forms of tourism and does not necessarily presuppose large absolute numbers of visitors.
Thus, relevant questions to be asked in analyzing a presumed scenario of overtourism include not only how many tourists visit the destination, but also what kind of tourism is prevalent (e.g., package tourism, individual tourism, cultural tourism, cruise tourism). How do tourists arrive (e.g., by airplane, by car, by train)? How do tourists behave locally (e.g., in terms of disturbances and intrusiveness)? Where do they stay (e.g., in hotels, in private apartments)? Moreover, when it comes to judging whether the social and ecological costs of tourism are economically worth it [1] (pp. [4][5]: How much value added stays in the destination and how much of it flows outside the region? These aspects demonstrate that the phenomena discussed under the term of overtourism, insofar as they relate to tourism at all, cannot be isolated from a destination's carrying capacity, which can include measurable dimensions, dynamic attributes, and constructions or perceptions [24] (pp. [4][5][6]. Indeed, carrying capacity [16,24,37] can significantly vary between destinations, thus explaining why mass tourism does not necessarily imply overtourism problems, as some destinations can absorb far higher numbers of tourist arrivals and overnight stays than others [2] (p. 9). Following Middleton and Hawkins, carrying capacity can be defined as a destination's "tolerance ( . . . ) to tourist activity and the limit beyond which an area may suffer from the adverse impacts of tourism" [38] (p. 239). Defined in this way, carrying capacity appears as the missing link in defining overtourism, at least if carrying capacity is understood in a broader, not exclusively quantitative sense by addressing the critique expressed about the concept [2] (pp. 2-3) [39]. Although terms such as tolerance and limits are relative, subjective, call for an interpretation, and depend on each individual destination with its specific characteristics, it seems conceptually plausible to understand overtourism as a scenario where a destination's tolerance for tourism is exceeded and the limits to harmful impacts are breached by the quantitative and qualitative manifestations of tourism found there.
Notwithstanding the lack of conceptual clarity both the term overtourism and the wider debate suffer from, the remainder of this paper applies a working definition of an overtourism scenario, understood as the consequences of tourism development exceeding a destination's carrying capacity.
In this understanding, overtourism is not simply a matter of too high numbers of tourist arrivals or overnight stays in a given destination, but also refers to patterns of tourism that exceed the economic, spatial, cultural, social, psychological, ecological, or infrastructure-related dimension(s) of a destination [2] (p. 5) [16] (pp. 1-2) [28] (p. 6) [38] (pp. 239-240) quantitatively, qualitatively, or both. Thus, overtourism as understood here refers not just to quantitative indicators of tourism but equally to the kinds of tourism that are prevalent, to the economic and social profile of the tourists a destination attracts, and to tourist behavior both before and during their visit to the destination [2] (pp. 2-3) [11] (p. 239) [37] (p. 279) [1,[39][40][41].
If an overtourism scenario is marked by some quantitative and/or qualitative manifestations of tourism development exceeding a destination's carrying capacity, we might expect the ensuing phenomena and problems to affect the evolution of a destination's tourism sector specifically and of the tourist destination generally. Evolutionary accounts such as lifecycle patterns or trajectories of growth and decline thus offer a suitable lens for analyzing the impact of an overtourism scenario on a destination and its local tourism sector. The next section therefore develops an evolutionary perspective on tourism development as a framework to examine the effects of overtourism on a destination's tourism sector.

Evolutionary Perspectives on Tourism Development and Overtourism
Butler's tourist area lifecycle model [28] is arguably the most well-known classic account of the evolution of tourist destinations. When it comes to the latter stages of the cycle, Butler's tourist area lifecycle includes different possibilities of development: rejuvenation, stagnation, and decline [28] (pp. 7-9). Due to these diverse possibilities, the tourist area lifecycle somewhat departs from the idea of a deterministic lifecycle pattern and includes a certain degree of the analytical propositions of contextuality, path dependence and contingency postulated by Bathelt and Glückler [42] (p. 128). Nevertheless, the trajectories of rejuvenation, stagnation, and decline offered by Butler [28] (pp. 7-9) remain rather vague and may be oversimplified. If, as in the perspective followed here, the impact of tourism development cannot be limited to tourism's quantitative prevalence, but has to take into account qualitative effects of tourism on a destination, we might expect the trajectories to differ much more than in Butler's [28] model.
There are a few studies that attempt a further conceptualization of what happens during the latter stages of Butler's tourist area lifecycle [28]. On a conceptual level, Ma and Hassink propose a path dependence model for tourist destinations that includes trajectories of adaptation, stagnation, and decline, with the latter related either to external shocks or to internal lock-in [26] (pp. 96-99) [27]. Knowles and Curtis describe how mass tourism resorts in England and Spain developed after stagnation [43] and propose a quasi-deterministic trajectory these mass tourism resorts might follow. Priestley and Mundet illustrate processes of decline and rejuvenation in mass tourism resorts in Catalonia [44]. Agarwal calls for a more refined conceptualization of the tourist area lifecycle's latter stages by arguing that the permanent reorientation of tourist destinations is a necessity [45] (pp. 71-72) and presents strategies employed during these latter stages under the theoretical framework of the restructuring thesis [32]. However, such a perspective tends to mix the strategies of firms, collective agents, or destination managers and policymakers, and thus does not sufficiently distinguish between restructuring trajectories at the firm level, at the sectoral level, or at the destination level. For instance, some of the restructuring strategies proposed by Agarwal [32] (p. 37) might coexist and together steer a destination's local tourism sector in a certain direction. Thus, a more refined conceptualization of how the local tourism sector evolves in a destination from a collective, sectoral, and spatial perspective is needed.
Following Ma and Hassink [26], such a refined conceptualization can build on a combination of Butler's tourist area lifecycle [28] and evolutionary economic geography. To further understand processes of the growth or decline of an industry within a local or regional economy, the regional industrial path development approach of evolutionary economic geography is useful [29,[46][47][48][49][50]. The approach offers a complementary, but more elaborate, perspective on the trajectories regional or local industries might embark on. Following Blažek et al. [29], Grillitsch et al. [46], and Isaksen et al. [47,48], six types of positive path development (path extension, path branching, path diversification, path creation, path importation, and path upgrading) and four forms of negative path development (path contraction, path downgrading, path dislocation, path disappearance) can be identified [29,[46][47][48][49][50]. However, the literature on path development so far does not pay specific attention to tourism. Therefore, the peculiarities of the tourism sector are not yet addressed in this strand of research. Still, with some qualifications, the path development approach can help us to reach a more detailed understanding of the possible sectoral trajectories in tourism. For a destination suffering from an overtourism scenario, the most relevant trajectories for the tourism sector to take are path branching or path diversification by developing new niche offers apart from quantitative growth in visitor numbers [1] (p. 6), path importation by developing new offers through investment from outside the destination, or path upgrading by moving into higher value-added and/or more sustainable forms of tourism [46][47][48]. These forms can be grouped under Butler's rejuvenation trajectory [28] (p. 9). Path creation is an unlikely option at the end of the tourist area lifecycle because it does not lead to the rejuvenation of an existing tourism sector, but refers to the initiation of a completely new industry, probably outside the tourism sector. Still, path creation can happen, for example if the tourism sector embarks on a trajectory of decline or even ultimate path disappearance. However, in this case, path creation would refer to another industry or sector, thus analytically not being part of the tourist area lifecycle. Butler's stagnation trajectory [28] (pp. [8][9] roughly corresponds to what the path development literature calls path extension [46][47][48]. The forms of negative path decline, i.e., path contraction, path downgrading, path delocalization, and ultimate path disappearance [29], provide more specifics on how precisely Butler's "decline stage" [28] (p. 9) might look. Table 1 summarizes how the forms of path development identified in evolutionary economic geography [29,[46][47][48][49][50] relate to the trajectories at the end of Butler's tourist area lifecycle [28].  [47,48], and Tödtling and Trippl [50].
The forms of path decline listed in the third row of Table 1 are where the remainder of this paper will focus, asking the question: How precisely does an overtourism scenario relate to the precise pathways of decline of a destination's tourism sector, provided that exogenous change due to policy interventions or public pressure does not occur? This question exhibits high practical relevance because the point where the danger for decline sets in is what Knowles and Curtis regard as "perhaps the most crucial point of action" [43] (p. 88).
As Butler's tourist area lifecycle relates to the limits of a destination's carrying capacity as the source of eventual decline [28] (p. 6), overtourism (as understood here) is a driving force for the latter stages of the lifecycle if rejuvenation does not ensue, probably due to measures either to attenuate the impact of tourism or to extend a destination's carrying capacity. This is when, in Butler's words, the tourism development in a destination enters the "critical range of elements of capacity" [28] (p. 7). Butler assumes that rising tourist numbers render a destination increasingly unattractive [28] (p. 10), but a purely quantitative view does not capture the full picture of the phenomena discussed in the overtourism debate. Instead, qualitative factors of overtourism need to be considered, e.g., visitor behavior that may depend in part on the kind of tourism promoted, such as the "beer bike" example [2] (p. 6) illustrates.
It is plausible that in the best case, policy interventions may, if successful in limiting tourism development and stopping it from exceeding a destination's carrying capacity, contribute to Butler's rejuvenation trajectory [28] (pp. 9-10) or to path upgrading [46][47][48]. Indeed, rejuvenation can come from within the tourism sector, as is the case when new attractions are built or new natural resources are harnessed [28] (pp. 9-10). However, such a trajectory seems unlikely in an overtourism scenario because adding a new tourist attraction or harnessing new natural resources would only exacerbate the underlying problems. Therefore, in an overtourism scenario rejuvenation or path upgrading is plausible either through exogenous changes to the tourism sector and, notably, policy interventions or through the local tourism sector changing its profile by shifting towards more sustainable and higher value-added forms of tourism. The latter case, however, is unlikely to happen unless incentives for doing so are in place, again calling for exogenous changes. Alternatively, incentives for path upgrading may emerge due to the path decline of the tourism model hitherto pursued, thus presupposing a prior trajectory of path contraction, path downgrading, or path delocalization.
Another plausible alternative is that Butler's stagnation trajectory [28] (p. [8][9] or path extension [46][47][48] can lead to overtourism phenomena simply continuing. If one assumes that the phenomenon Higgins-Desbiolles calls "tourism killing tourism" [9] (p. 157) does not occur, such a trajectory would be economically feasible, although one might assume that public pressure or "tourismphobia" [10,13] would mount due to the wider socio-ecological consequences of overtourism [3] such as overcrowding, touristification, rising real-estate prices, or gentrification [1] (pp. 3-4) [2] (p. 7) [19]. This rising public pressure makes a simple continuation of a pathway marked by overtourism unlikely, albeit for reasons outside the interior economic logic of the tourism sector. Indeed, this is what seems to be happening in Barcelona [11].
By looking at the inner dynamics of the tourism sector in the absence of exterior impulses for change such as policy interventions or wider public pressure, the remainder of this paper focuses on Butler's scenario of decline [28] (p. 9) and, more precisely, on forms of path decline as defined by Blažek et al. [29], i.e., path contraction, path downgrading, path delocalization, and path disappearance, and discusses how these forms of path decline may be related to overtourism, understood as the consequences of tourism development exceeding a destination's carrying capacity.
As the four forms of path decline demonstrate, path decline does not have to mean absolute path disappearance. Indeed, complete disappearance of tourism from a destination may be plausible under the assumption that overtourism is a self-limiting phenomenon, even in the absence of exogenous changes due to policy intervention or public pressure. Alternatively, overtourism could be a cumulative phenomenon, at least as long as further growth of the tourism sector in a destination does not face limits set by the wider societal, ecological, or political system. Whether overtourism is a self-limiting phenomenon or a cumulative one is an empirical question that can be answered only in the long term, if at all. So far, it does not seem to be self-limiting [1] (p. 4), as the long-term growth of one of the main examples of the overtourism debate, Barcelona, demonstrates [7] (p. 8). On the other hand, it is hard to imagine how tourism development exceeding a destination's carrying capacity could possibly not lead to a degradation of a destination's touristic value. After all, the World Tourism Organization's definition of overtourism explicitly refers to the degradation of the tourist experience as a crucial element of the phenomenon [6] (p. 4). The answer to this apparent paradox arguably lies in the multidimensional nature of overtourism [2] (p. 5) and notably in the fact that overtourism is not only a purely quantitative phenomenon, but is also, maybe even more so, a qualitative one. Even if tourist numbers continue to grow, a destination affected by phenomena related to an overtourism scenario can decline qualitatively, for example in terms of the quality of the tourism experience it offers or in terms of the value added it captures. This is why path contraction, path downgrading, and (partial) path delocalization offer scenarios of lock-in [25][26][27] into suboptimal trajectories, even without eventual path disappearance. This is why the remainder of this article is based on the assumption that an overtourism scenario is related to various forms of path decline of the tourism sector in affected destinations. The precise forms of path decline are at the core of the arguments in the next section.

Path Decline in the Context of Overtourism: A Typology of Scenarios
In terms of path decline that roughly corresponds to Butler's broad trajectory of decline [28] (p. 9), Blažek et al. [29] propose a taxonomy of three types of negative path development, all of them possibly leading to a fourth one. Path downgrading refers to processes that make a sector gradually move into lower value-added parts of the value chain. Under a trajectory of path contraction, a sector gradually becomes overly specialized. Path delocalization leads to the relocation of a sector to another region. Each of these three types of trajectories in extremis can result in the ultimate form of path decline, path disappearance, meaning that an industry completely vanishes from its established location [29].
The arguments made in this section are based on the assumption that absent exogenous changes as a result of policy intervention or public pressure, an overtourism scenario marked by tourism patterns quantitatively or qualitatively exceeding a destination's carrying capacity is likely to lead to the deterioration of the competitive position of a destination's tourism sector. Under this assumption, each of the three forms of path decline can be driven by problems of overtourism reinforcing themselves over time, and the fourth, ultimate form of path disappearance might be the last, but still fairly hypothetical, possibility. Table 2 provides an overview of the four forms of path decline proposed by Blažek et al. [29], i.e., path downgrading, path contraction, path delocalization, and eventual path disappearance, and discusses the possible impact of an overtourism scenario as understood here, i.e., the consequences of tourism development exceeding a destination's carrying capacity. Furthermore, Table 2 gives some indicative examples of destinations that might risk embarking on a specific form of path decline, respectively, in the absence of major exogenous changes to the tourism sector such as substantial policy interventions, considerable public pressure, or both. Table 2. Forms of path decline in the context of overtourism.

Path contraction
Increasing focus on standardized forms of tourism (e.g., Fordist sun, sand and beach mass tourism, bus tourism, cruise tourism, reliance on low-cost carriers) • Palma de Mallorca [19,51,52] Path downgrading Growth of tourism segments with lower value added and intense behavior-related (e.g., noise, drinking), cultural (e.g., touristification), or ecological (e.g., degradation of the landscape) disturbances, displacing higher value-added and/or possibly more sustainable tourism segments (e.g., cultural tourism, eco-tourism) • Venice [3,5,12,15,16] • Barcelona [3,7,[11][12][13] Path delocalization Deterioration of a destination's tourism infrastructure (e.g., lack of investment), possibly due to the displacement of higher value-added segments, and shift of investments to another location • Possibly Dubrovnik [4,14] with eventual delocalization to Kotor [53] Path disappearance Theoretical possibility of one of the three forms ultimately leading to the complete vanishing of the tourism sector from a destination, but unlikely
Path contraction may be a consequence of the increasing focus of an overtouristed destination's tourism sector on standardized forms of tourism such as Fordist-style sun, sand and beach mass tourism [32,33], bus or cruise tourism [15,18], or a model geared towards attracting low-cost carriers to generate tourism volume [17]. Palma de Mallorca can serve as an example for a destination that has focused on mass tourism that seems to increasingly permeate the city itself [19,51,52], thus risking a pattern of overspecialization and a loss of diversification characteristic for path contraction [29] (p. 9).
Path downgrading can be driven by the growth of tourism segments with lower value added and intense behavior-related, cultural, or ecological disturbances. These disturbances include noise, drinking, and other problematic behavior by visitors, touristification of inner cities and shopping districts, or emissions, waste, and the degradation of the landscape [1] (p. 4) [2] (p. 5). As a result, higher value-added and/or possibly more sustainable tourism segments such as cultural tourism or eco-tourism might be displaced. Venice [3,5,12,15,16] and Barcelona [3,7,[11][12][13] can serve as examples for such a downgrading process, at least in their most touristed historical areas. Given the prominence of these two examples and the clear link to problems commonly discussed in the overtourism debate, a plausible hypothesis to be empirically tested in further research is that path downgrading could be the most prevalent trajectory currently experienced by the tourism sector in destinations affected by an overtourism scenario.
Path delocalization is more difficult to imagine in the tourism sector than for manufacturing industries such as automobiles or textiles or service industries such as back-office activities [29] (pp. 9, 11) since tourism is, at least to a considerable degree, place-bound. Still, partial delocalization can occur if higher value-added segments of the tourism sector are displaced from a destination under an overtourism scenario, possibly as the consequence of earlier path contraction or path downgrading. This case, then, demonstrates how pathways can shift and lead to one another. A deterioration of investments in the segment can be the result of an overtourism scenario, while investment opportunities are being looked for in other, alternative destinations. In this way, the partial delocalization of the tourism sector from an overtouristed destination to another destination with sufficient carrying capacity can occur. Examples are difficult to find but plausible candidates for such a scenario can be identified. For instance, if the overtourism problems found in Dubrovnik [4,14] are not resolved, a delocalization of higher value-added tourism segments to comparable destinations such as Montenegro's Kotor [53] could ensue.
There is not (yet) any visible example for path disappearance of a destination's tourism sector after an overtourism scenario known to the author, although Venice [3,5,12,15,16] can be thought of as the one that might eventually come closest. Indeed, this trajectory is rather hypothetical because there is no deterministic course from any of the other three pathways to ultimate path disappearance. Furthermore, complete path disappearance is unlikely in the first place because some form of tourism is likely to stay in any case. Any of the other three forms of path decline-path contraction, path downgrading, and (partial) path delocalization-can be thought to lead to some form of stability, albeit one that entrenches the overtourism scenario through lock-in [25][26][27], and one that is marked by wider societal or ecological sustainability deficits that make exogenous change through policy interventions or rising public pressure likely. Therefore, it is not surprising that complete path disappearance is a fairly theoretical possibility in tourism, even in an overtourism scenario. Still, the possibility cannot totally be ruled out and demonstrates the ultimate dangers an overtourism scenario can pose to a destination's tourism sector by eventually bringing about the phenomenon of "tourism killing tourism" [9] (p. 157).
The possible examples of destinations at risk listed in Table 2 demonstrate that a destination possibly embarking on one of the forms of path decline does not necessarily need to witness decreasing numbers of tourist arrivals, as Barcelona's visitor boom in recent years illustrates [7] (p. 8). The examples listed should be seen as indicative, and in-depth empirical research will be needed to verify whether these destinations are indeed on a given trajectory of path decline in the absence of exogenous change such as policy interventions-which is a rather hypothetical scenario, because some form of exogenous change through at least limited policy intervention is likely to happen, given the wider societal and ecological sustainability deficits of an overtourism scenario. The argument put forward here is that these destinations may be at risk of such a trajectory, which might eventually lead to (at least partial) path disappearance if the tourism sector is left on its own without effective policy intervention. Still, path disappearance is not the necessary end of either path contraction, path downgrading, or path delocalization trajectories. In contrast to Butler's [28] focus on the development of visitor numbers in distinguishing between his rejuvenation, stagnation, or decline trajectories, the taxonomy of path decline followed here implies that even if pure visitor numbers stagnate or further increase, path decline can continue and reinforce itself, e.g., through path downgrading which may even be the most widespread case. Path contraction does not necessarily imply contracting visitor numbers but suggests overly specialized forms of tourism and a lack of product diversification which may lead to strong seasonality, the decreasing duration of stays, or both. Under path downgrading, a local tourism sector can, by losing its key competitive advantages, trade itself down into low-value added, low-quality, and/or low-sustainability activities. Path delocalization may indeed lead to decreasing visitor numbers but can be partial, e.g., by displacing higher value-added and possibly more sustainable tourism segments to other destinations while keeping the lower value-added and less sustainable ones in the destination at hand [29] (p. 9). This is not to claim a general correlation between value added and sustainability, because there are segments characterized by high value added but low sustainability, e.g., in ecological terms [36,51]. If, however, segments that are marked by both high value added and low sustainability are displaced, the pattern of resulting (partial) path delocalization is most serious.
As the discussion above demonstrates, the types of path decline listed are not mutually exclusive and perfectly clear-cut, as destinations can witness patterns that can be classified under more than one type [29] (p. 6). In this sense, the typology proposed by Blažek et al. [29] is stylized and refers to dominant tendencies. Furthermore, it is possible that a destination's tourism sector initially embarks on one trajectory of decline but then switches to another. Finally, turning around the development by embarking on a positive form of path development such as path upgrading, path branching, or path diversification [46][47][48] is possible but will often need external change to happen.
The argument made here is, to a certain degree, hypothetical because in each case, exogenous change due to policy interventions or rising public pressure can reduce the risks of path decline and notably of ultimate path disappearance. Whether this kind of exogenous change will happen remains to be seen in each specific case. Nevertheless, the possibility of affected destinations embarking on a trajectory of path decline should be taken seriously by policymakers and tourism sector stakeholders. Taking this possibility seriously means considering effective measures of tourism policy and destination management. The next section proposes some preliminary conclusions for tourism policymakers and destination managers to draw.

Conclusions for Tourism Policy and Destination Management
While this article has attempted to shed light on processes of the possible decline of an overtouristed destination's tourism sector, the obvious policy challenge is how to prevent decline. While one might argue that, from an ecological and social perspective, for some destinations with sensitive ecosystems, a declining tourism sector could actually prove beneficial for society, such a course is not desirable from the point of view of tourism stakeholders and thus is politically unlikely to be accepted. A moderate position that seeks to balance the economic, ecological, and social needs of citizens and stakeholders in a destination, from within the tourism sector and in society as a whole, would be to achieve a course of rejuvenation [28] (p. 9) not in terms of visitor numbers, but in terms of socio-economic benefits, notably through some form of path upgrading, path branching, or path diversification [46][47][48]. Calling for the upgrading a destination's local tourism sector is not to deny the need to reorient the growth model of global tourism that is still largely prevalent towards a model of mobility embedded into wider societal and ecological questions of sustainability [9]. On the contrary, if we assume that overtourism scenarios lead to societally and ecologically unsustainable forms of tourism further entrenching and reinforcing themselves according to some of the forms of path decline studied (at least in the absence of eventual complete path disappearance), the need for a destination's tourism sector to upgrade or diversify towards a model that balances economic interests with wider societal and ecological ones [1] (pp. [4][5] becomes even more pressing. Within the current overtourism debate, there is an urgent need for more research to identify options for tourism policy and destination management on how to steer the tourism sector in destinations affected by an overtourism scenario away from trajectories of path decline. A limited body of literature discusses possibilities to address the latter stages of Butler's tourist area lifecycle [28], but these studies do not provide immediate answers for destinations affected by overtourism scenarios. For example, Agarwal categorizes several restructuring strategies that can contribute to a destination's renewal, such as improving product or environmental quality, specializing or diversifying the offer, or repositioning a tourist destination [32] (pp. [36][37][38][39]. However, how these strategies relate to different trajectories of decline and to overtourism scenarios remains unclear. Specifically, what Agarwal [32] (p. 37) calls "repositioning" could prove particularly relevant for overtourism scenarios but still relates mostly to classic mass tourism. Similarly, Knowles and Curtis make an attempt to explore possibilities to prevent the decline of Spanish mass tourism destinations [43] but their arguments, too, do not provide deeper insights into how overtourism phenomena shape these scenarios, and, critically, they center around the dominance of tour operators in the classic form of Mediterranean mass tourism.
Furthermore, there are studies on the upgrading of mass tourism destinations such as on Malta's Bugibba [54], Mallorca [51,52], or beach resorts in Catalonia [44]. These studies refer to measures such as converting three-star hotels into upmarket hotels [44,54]. The transferability of these studies' outcomes to overtourism scenarios is not to be taken for granted, since overtourism is a phenomenon different from mass tourism [2] (p. 9) and "typical" Mediterranean mass tourism destinations such as Mallorca's [51,52], Malta's [54], or Tunisia's [33] coastal tourism zones tend to be small, sharply delimited tourism agglomerations [35] (p. 1) [36] (p. 39). Furthermore, Mediterranean mass tourism is marked by the dominance of tour operators [43] (p. 94), while overtourism is a more complex phenomenon. Nevertheless, depending on the destination at hand, both phenomena are not isolated from each other. The example of Palma de Mallorca's city center [19] shows that mass tourism can spill over to other areas and lead to overtourism problems. More generally, even if these spillovers do not occur, the longer-term experience with upgrading mass tourism destinations, including failures [e.g., 36,51], could prove useful for designing upgrading strategies for overtouristed destinations. In particular, it should be considered that upgrading does not necessarily mean "luxury" tourism which is not in and of itself sustainable [1] (p. 10) [36]. Furthermore, in contrast to the decline of Mediterranean Fordist mass tourism resorts, addressing the possible decline of the tourism sector in destinations suffering from an overtourism scenario will require more than upgrading the quality of the offer. Instead, upgrading or diversifying the tourism sector in overtouristed destinations is a comprehensive task that requires the development of a different model of tourism that is less reliant on large numbers of visitors or forms of tourism that bring with them severe disturbances for the local population and environment. Such a different model could involve, for instance, niches marked by a better ratio between local value added and socio-ecological footprints such as agritourism or culinary tourism [1] (p. 6).
Despite its context (Mediterranean Fordist mass tourism), the example of Catalonia's Torroella de Montgrí/l'Estartit, which experienced diversification into environmentally oriented forms of tourism [44] (pp. 97-100), could provide some guidance for structurally comparable destinations suffering from an overtourism scenario.
Organizing an effort to steer a tourist destination away from a trajectory of path decline, notably by shifting towards a trajectory of path upgrading or diversification, will require widespread collaboration between the public and private sector, for instance in the form of destination management organizations (DMOs) [55], cluster initiatives [35,56] or wider, cross-sectoral collaborative schemes such as smart specialization strategies [35], tourism improvement districts [57], or other fora that enable the discovery and development of solutions that both serve the upgrading of a destination's tourism sector itself on the one hand and ecological and societal needs articulated by civil society agents on the other hand [1] (pp. 7-8) [2] (p. 10) [24] (pp. 5-6) [58].
Due to the multidimensionality of overtourism phenomena and the need to find solutions adapted to the local context of the destination at hand [2] (pp. 1, 5, 10), the precise measures for DMOs and individual firms from the tourism sector will vary from destination to destination, depending on the precise problems that are prevalent. Thus, a collective agenda to shift from a trajectory of decline to a trajectory of path upgrading will have to consider aspects of policies (going beyond the activities of the tourism sector), institutions, organization, and visitor behavior [1].

Avenues for Further Research
The present article has sought to present a conceptual basis for better understanding the trajectories of decline of the tourism sector in a destination affected by an overtourism scenario. Based on Butler's classic tourist area lifecycle model [28], the article has shown the usefulness of a more detailed conceptualization of path decline in tourism by applying the path development approach known from evolutionary economic geography [29,[46][47][48][49][50].
Further research is needed to advance our understanding of path decline in tourism. Conceptually, further refining the trajectories of path decline by looking at the co-evolution between the tourism industry, the tourism product, and tourism-related institutions as suggested by Ma and Hassink [26] could prove promising to better understand what drives the trajectories of path decline. In-depth empirical research in destinations affected by overtourism scenarios will be useful to gain more detailed insights into the workings of path decline, including their conditions, and possibilities for a destination's tourism sector to shift from a trajectory of decline towards a positive form of path development. The indicative examples of destinations at risk offered in the discussion above are meant to provide an impulse for further discussion only, notably because they illustrate cases where a trajectory of decline might ensue in the absence of exogenous change. Still, how the local tourism sector in these destinations at risk actually develops is a factual question that should be addressed empirically in further research. Additional cases might yield more comprehensive insights and it is possible that other, less well-researched destinations provide useful examples for actual instances of path decline that have led to the ultimate form of decline, path disappearance, or to the successful rejuvenation of the destination's tourism sector by shifting towards a positive form of path development such as path upgrading or diversification. Furthermore, it will be interesting to examine what role the institutional context [59], institutional change [60], and institutional entrepreneurship [61] can play to drive changes in path development in a destination's local tourism sector. These questions will be of particular relevance in cases of a local tourism sector having managed to shift its course from a trajectory of path decline to one of path upgrading and, more generally, how institutional arrangements [62] prevalent in a destination's local tourism sector affect path development [63]. Drawing on pioneering studies that address the role of institutions in developing more sustainable forms of tourism [64] could inspire directions for empirically founded research that will be needed.
However, the present article's arguments suggest that such an upgrading process is unlikely to happen in a scenario of overtourism if an affected destination's tourism sector is left on its own. Exogenous events such as public pressure or policy interventions will often be necessary to break the spirals of decline laid out in this article. Precisely which policy interventions are effective in breaking these spirals and in instigating an upgrading process is the crucial question that will have to be addressed in further, policy-relevant research. For now, understanding why and how not only a destination as such, but also, more specifically, its local tourism sector embarks on a trajectory of path decline because of the harmful consequences of an overtourism scenario is important and should serve as a call for further research within the academic community and as a call to action for policymaking communities to find and implement appropriate solutions.