Annals of Tourism Research, Vol. 30, No. 4, pp. 886­905, 2003 2003 Elsevier Ltd. All rights reserved Printed in Great Britain 0160-7383/$30.00 www.elsevier.com/locate/atoures doi:10.1016/S0160-7383(03)00061-6 TERRITORIAL PLANNING AND TOURISM DEVELOPMENT TAX Claudio A.G. Piga Nottingham University, UK Abstract: An economic model of land taxation involving a local government and a private developer constitutes the theoretical framework in this research. The model hinges around a two-tier approach including both a conservation and an efficiency criterion. The analysis indicates that sustainable tourism calls for the use of land taxation and planning legislation simultaneously geared to the achievement of efficiency and the signaling of the government's commitment to conservation policies. To provide support for the theoretical conclusions, an Italian case study is discussed, showing how the local government chose not to yield to a developer's requests by not changing a norm prohibiting construction near of the coastline. Keywords: land usage, sustainability, game theory, Sardinia. 2003 Elsevier Ltd. All rights reserved. Résumé: Aménagement du territoire et développement du tourisme. Un mode`le écono- mique des impo^ts sur la propriété foncie`re, basé sur la participation du gouvernement local et d'un promoteur de construction, constitue le cadre théorique de cette recherche. Le mode`le dépend d'une approche a` deux niveaux qui inclut des crite`res de la défense de l'environnement et de l'efficacité. L'analyse indique que le tourisme durable exige l'utilis- ation des impo^ts sur la propriété foncie`re et des lois de planification qui contribuent a` la réalisation de l'efficacité et qui signalent l'engagement du gouvernement a` une politique de défense de l'environnement. Pour soutenir les conclusions théoriques, on discute d'une étude de cas italienne qui montre comment le gouvernement local a choisi de ne pas céder aux demandes d'un promoteur en ne changeant pas une norme qui interdisait la construc- tion pre`s de la co^te. Mots-clés: utilisation de la propriété foncie`re, durabilité, théorie du jeu, Sardaigne. 2003 Elsevier Ltd. All rights reserved. INTRODUCTION Environmental quality is often a key success factor for tourism. At the same time, tourism makes extensive use of natural resources, thereby jeopardizing its long-term viability. To address such a dilemma, from the general notion of sustainable development (SD) tourism scholars have coined the term "sustainable tourism" (ST) that encompasses a set of principles, business methods, and policy prescriptions relevant to the tourism industry (Sinclair and Stabler 1997). However, recent research has advocated the need to reconcile the concept of ST with Claudio Piga is Lecturer in Economics at the Nottingham University Business School (Nottingham NG8 1BB, UK. Email ) and a collaborator with the DeHaan Travel and Tourism Research Institute. His research aims at studying the tourism industry by combining theoretical contributions from such areas of economics as industrial organization, game theory and environmental economics. He is also doing research on low cost airlines. 886 887CLAUDIO PIGA that of SD, by making the concerns of the former adhere more strictly to the tenets of the latter (Collins 1999; Hunter 1997). This paper looks at the issues of tourism development and environmental conser- vation through the lens of SD principles and assesses the pivotal role that local governments can play in designing policies that make the two perspectives compatible. A common denominator in the literature on SD is the Bruntland Commission's definition that "SD is development that meets the need of the present without compromising the ability of future generations to meet their own needs" (WCED 1987). Such a mandate affirms the importance of intragenerational and intergenerational equity issues. As far as the former is concerned, "SD places emphasis on providing for the needs of the least advantaged in society" (Pearce, Markandya and Barbier 1989:2). This is particularly relevant in developing coun- tries, where the need to generate income is more likely to lead to a rapid exploitation of the resource base and an uneven distribution of the related profits between foreign investors and host populations. To prevent this, it is generally recommended that local populations and governments be directly involved in the shaping of tourism develop- ment activities in association with foreign developers, because locals can better assess the short and the long-term effects of growth. As far as the latter is concerned, intergenerational fairness implies that future generations should receive a fair share of the net benefits supplied by tourism (Hunter and Green 1995). The fact that development is sustainable only as long as future generations are fairly treated led to the view that ...we in the present generation are but the present tenants of the earth, not its absolute owners. As present tenants we have the right to make use of its productivity, but not the right to impair its pro- ductivity for its later tenants--the future generations" (Page 1997:586). In theory, future generations are entitled to just compensation for the current actions that lead to a depletion of natural resource. In practice, doing so requires making difficult judgments about the substi- tutability of natural capital with physical capital. This is a crucial point in the case of tourism, whose activity relies extensively on the trans- formation of natural capital into accommodation and service facilities. Thus, this industry faces the particularly difficult challenge to "develop tourism capacity and the quality of its products without adversely affect- ing the physical and human environment that sustains and nurtures them" (Cronin 1990:13). It is argued, however, that the WCED definition is too vague to use as a working tool, and often results in ambiguous admonition for policymakers (Norton and Toman 1997). Indeed, the requirement that the needs of the present be met without compromising the ability of future generations to meet their needs, could be satisfied simply by allowing constant consumption over time at no more than a subsist- ence level. Although such a policy would achieve equity among differ- ent generations, it could hardly be accepted as a reasonable target for 888 SUSTAINABILITY AND LAND TAXATION public policy, as it may fail to incorporate a notion of dynamic efficiency. In the following analysis, efficiency implies the notion of "non-wastefulness", that is, the possibility that inputs are transformed in a way that leads to an increase in individuals' well-being in a sus- tainable manner (including consumption above the subsistence level for every generation). Such an increase in individuals' welfare entails a decision on the optimal allocation of scarce resources (such as natu- ral assets) among alternative uses (exploitation or preservation). Thus, the real challenge is to indicate sustainable policies that combine intragenerational and intergenerational equity with efficiency con- siderations in a mutually compatible manner, taking into account that tourism can never be totally without environmental impacts (Page 1997). As far as intergenerational equity and efficiency are concerned, assessing under what circumstances turning natural resources into capital stock (a hotel) leads to an increase in the well-being of the present and future generations is central. Furthermore, the notion of efficiency is closely linked with that of intragenerational equity when the local populations of the tourism areas are allowed to appropriate a fair share of the net benefits generated by tourism. The theoretical analysis in this article presents a policy measure that enables local governments to retain the surplus from tourism activities when these are run by foreign organizations. The existing literature has refined the definition of sustainable development--from very weak to very strong--depending on the importance given to such notions as reversibility and substitutability between physical and natural capital (Collins 1999; Howarth 1997; Hunter 1997; Norton and Toman 1997; Page 1997). Unfortunately, the interpretation of such concepts varies, depending on the disciplinary approach being adopted. For instance, economists and ecologists assign different meanings to the notions of substitutability and reversi- bility. For economists, substitutability refers to the possibility of main- taining a desired level of production using different combinations of inputs, while reversibility indicates the extent to which a given resource becomes scarce as a consequence of human actions. For ecologists, these two factors determine such ecosystems' physical properties as, for instance, resilience. In this sense, a more resilient ecosystem is one that is more likely to revert to its original condition after a perturbation and/or to find other substitutes in the event that one of the ecosystem attributes is diminished. Applying the notion of weak sustainability is possible if it is assumed that natural and physical capital are substitutable, or when changes to the natural asset base are reversible. This implies that under the weak sustainability paradigm, intragenerational equity is obtained through the distribution of the efficiency gains arising from the implemen- tation of development projects, while future generations are compen- sated for the loss of natural assets by inheriting a greater stock of physi- cal capital. However, even advocates of weak sustainability acknowledge that under certain circumstances (such as when physical capital is a poor substitute for the natural resource), efficiency considerations ought not to occupy a central role. For instance, it has been argued 889CLAUDIO PIGA that one generation might set aside special places and features such as the Grand Canyon or the Reef Barrier for future generations because of their intrinsic qualities. The uncertainty surrounding both the effects of human intervention on the environment and the likelihood of finding feasible technologi- cal solutions to environmental problems provides an important argu- ment for strong sustainability. By this it is meant that the opportunities of future generations can be secured only if natural resources and environmental quality are specifically conserved for their benefit. Thus, within the strong sustainability paradigm, conservation concerns are paramount. Combining Conservation and Efficiency The foregoing discussion has illustrated that intragenerational/ efficiency and intergenerational equity goals are less likely to be in conflict when considered from within a weak sustainability perspective, while supporters of strong sustainability are inclined to reject any trade- off between conservation and efficiency. In line with the latter view, it has become abundantly clear that tourism cannot continue to be the killer of the "goose laying golden eggs" (Hunter and Green 1995: 42). However, it has been argued that strong sustainability, with its empha- sis on intergenerational fairness, can conflict with the efficient use of resources that may engender beneficial effects for both present and future (Hunter 1997). That is, the strong sustainability approach is often associated with an anti-economic growth position that appears to deny the world's poor the opportunity of meeting basic needs, both in the short and the long run. Therefore, this article takes the center ground stance indicated in Hunter (1997) by arguing that the essential role of ecological conservation in the implementation of sustainable tourism strategies should, depending on the circumstances, be comp- lemented by efficiency considerations. This is in line with many metho- dological contributions aimed at bringing together efficiency and intergenerational equity. The combination of a conservation with an efficiency criterion consti- tutes the central element of the two-tier method advocated by Page (1977). Using this approach, problems are categorized as being intrag- enerational or intergenerational in their effects, in the sense specified above. That is, when the latter issues figure prominently, more atten- tion is given to conservation, with possible applications of the "safe minimum standard of conservation principle", which places the bur- den of proof on today's resource allocators to demonstrate that their behavior is consistent with intergenerational fairness (Howarth 1997). In practice, this approach recognizes the fundamental right of future generations to inherit an intact stock of natural resources, unless the costs of foregoing the resource exploitation (that is, the loss of the efficiency gains from development) are unbearably high (Toman 1994). Page provides a list of inter-related issues that need to be addressed to achieve a satisfactory combination of efficiency and equity. First, 890 SUSTAINABILITY AND LAND TAXATION instruments need to be identified. He suggests that "shifting the tax base toward virgin material taxes and taxes on environmental harms would work toward sustainability" (1997:593). Second, intergener- ational equity should always come first when the resource is essential. He cites the US Drinking Water Regulation as an example of public intervention based on equity grounds, as future generations are entitled to a safe supply of drinking water. The example that comes to mind in the case of tourism is the establishment of parks or natural reserves, where regulatory measures are taken toward environmental protection so that the functional integrity of essential, and possibly unique, natural ecosystems is preserved as far as is possible, for the benefit of future generations. Indeed, once a law attributes the status of park or of natural reserve to a given geographical area, heavy restric- tions are imposed on the possibility of developing the area, that curtail or impede tourist activity. More importantly, to guarantee that the resource base remains intact, and can be bequeathed to future gener- ations, the status of park cannot be abolished and remains associated with the area indefinitely. Third, and related to the latter point, the role of the legal framework in which the decision-makers operate is crucial. Practically, this entails that the environmental decisions associated with important equity and efficiency considerations should be dealt with in a manner similar to that used to preserve a system of constitutional law. This is because the "framers of a constitution are expected to abstract themselves from their own narrow self-interest and establish the rules of the game that are sustainable indefinitely" (1997:590). It is essential for modern societal institutions to maintain the constitutional system: to protect it from myopic opportunism triggered by short-term benefits, it is com- monplace to have special procedures to modify constitutional dictates. By the same token, environmental problems with important intergen- erational equity aspects should be dealt with like constitutional issues whereby the future generations' entitlement to an intact resource base has paramount importance. Three points that are central in this study may be inferred from the previous discussion: the importance of natural resources, the identifi- cation of appropriate policy instruments, and the legislative frame- work. This article aims to clarify some of the relationships among these issues. At the same time it develops a conceptual framework encompassing all the basic elements that are taken into account in the public assessment of new tourism projects. To this purpose, a theoreti- cal economic model of land taxation is illustrated. Its results show the crucial role played by a tax on land use for the joint achievement of conservation and economic efficiency goals. In addition, its cost-benefit approach clearly indicates that in environmentally sensitive areas-- where the resource is essential--the tax should be set at a level that deters development. This result is thus equivalent to the creation of a park or a reserve that remains as close to its original form as possible. The importance of the legislative framework is investigated using the case study of the so-called "masterplan" in North Sardinia (Italy). The evidence presented, together with considerations from the theoretical 891CLAUDIO PIGA model, are used to shed light on the Sardinian government's refusal to grant a developer special exemptions from existing territorial planning legislation. Indeed, the Sardinian government decided to forego the short-term benefits of a large tourism project because the development allowed the possibility of construction within 300 meters of the coast- line, a practice that is prohibited by Sardinian law (article 10bis of law n.45 issued in 1989). The local government recognized that the defense of the conservation principle embodied in the regional law could not be a matter of bargaining, even though this would have entailed, as it actually did, the withdrawal of the project by the devel- oper. Regardless of the merit of the specific project and the good repu- tation of the developer for quality tourism, the Sardinian government's stance on the project aimed to ensure that other developments might not be created unless sustainable principles were applied (Collins 1999). SUSTAINABILITY AND PUBLIC INTERVENTION There is no formal economic model explicitly linking tourism and sustainable development. However, there exists a growing literature aimed at integrating sustainability and formal economic analysis. Fau- cheux, Pearce and Proops (1996) classified the existing analytical approaches to sustainability using four categories of models: neoclassi- cal, evolutionary, ecological economic, and neo-ricardian. The model presented in this article falls within the first category but departs from existing models by adopting a game theoretic approach where differ- ent incentives faced by public and private institutions are taken into account. A technical presentation of the model appears elsewhere (Piga 1999, 2003). The emphasis here will be the description of the assumptions used in this model, and the policy recommendations that can be drawn. This is amplified with a case study that complements and supports the theoretical analysis. Sustainability and Taxation The model under analysis is dynamic: it considers a sequence of time periods and how decisions taken in early periods influence the outcomes of subsequent periods. A dynamic approach allows a better identification of the project's long-term effects, notably on land, which is the natural resource under study. This is particularly relevant for tourism. For territorial planning, environmental design, and land use are crucial factors that create and sustain a tourism resort competitive advantage (Gunn 1994). At the same time, tourists increasingly expect a picturesque landscape to be integral to the holiday experience. More- over, land development is associated with various forms of environmen- tal costs. Both these aspects are captured in the following analysis. The first economic agent taken into account is a private tourism developer who owns a territory of size L. For instance, L could rep- resent the size of an island. The developer chooses the rate of land exploitation, that is, the portion of the site on which tourism facilities 892 SUSTAINABILITY AND LAND TAXATION will be erected. Denoting with B(t) the stock of developed land at time t and with (t) the size of territory on which the developer decides to build in period t, it is postulated that these two variables are linked by this simple law of variation over time: dB(t)/dt = (t). Such an expression indicates that in every period t a portion of land (t) is used to build accommodations. Thus, what is built this year is added to what was developed in previous years, thereby increasing the stock of buildings that constitute the site's capacity. The following example clarifies the meaning of the previous equation. Assume that at time t0 no land is developed, that is, B(t0) = 0. Then, after a given period of time (such as ten years): B(t0 + 10) = t = t0 + 10 t = t0 + 1 (t) = (1) + (2) + ... + (10). Obviously, the size of the development at time t, denoted by B(t), cannot exceed the total size of the territory, namely B(t) < L. The value of (t) may also differ from period to period. Another implication is that the transformation of land is irreversible: the site cannot be restored to its original condition. The discussion so far highlights the well known impact of tourism on environmental quality: "In reality, it is impossible to imagine any kind of tourism activity being developed and then operating without in some way reducing the quantity and/or quality of natural resources somewhere" (Hunter 1997:858). In this particular case, for tourism activity to take place, it is impossible to maintain the resource base intact (not to exploit land). It follows that a satisfactory combination of efficiency and conservation hinges around the identification of both the private and public benefits engendered by the development, of its environmental costs, and of the appropriate policy instruments that induce an optimal level of use of the natural resource. The first type of private benefit taken into account is the revenue derived from selling to tourists. This depends on the price of the hol- iday and on the site's capacity, B(t). In turn, it is assumed that the tourists' willingness to pay for a holiday (which represents the highest price they would pay) is positively influenced by the environmental quality of the site, which is measured by the amount of land that is left unused, namely L B(t) (Keane 1997). For simplicity, assume that the relationship between the site's environmental quality and willing- ness to pay, denoted by p(t), is linear: p(t) = a+L B(t), where the parameter "a" indicates the consumers' willingness to pay for all the other features, in addition to environmental quality, characterizing the holiday (gastronomy, entertainment, etc). More importantly, the pre- vious analytical expression indicates that the more the place is developed (the greater B(t) relative to L), the lower the price the devel- oper can charge for a holiday. Therefore, the developer, when decid- ing to expand (increase the size of B(t) through the choice of (t)), has to take into account that the revenue increase, due to the possibility of accommodating more tourists, may be more than offset by the reduction in price that is triggered by the deterioration in the environ- mental quality. Such a novel representation of the links between tour- ists' demand and environmental quality captures a peculiar feature of the tourism industry that is supported by empirical evidence. Font (2000) presents evidence supporting the notion that environmental 893CLAUDIO PIGA considerations are important drivers of tourism demand. Huybers and Bennett (2000) reach a similar conclusion in an analysis of changes in environmental quality and other features that affect demand. The local government is the second economic agent taken into con- sideration. The instrument used to achieve sustainability is a tax on each unit of newly built territory. In each period t, the total tax take is given by T(t) = (t)(t), where (t) denotes the land tax set by the local government. This clarifies the crucial role of the public sector: the tax is a cost to the developer and can be used by the government to appropriate some of the profits generated by the project. As dis- cussed below, the tax guarantees the achievement of intragenerational equity objectives, as the tax revenues in each year can be deployed to ameliorate the local public infrastructure which is then inherited by future generations. In line with the real-world situation where the legislative framework and the taxing policy is often a given for the developers, the economic model assumes that the local government has a first-mover advantage (that is, the government sets the tax level before the developer decides the expansion size). In the economic literature, the player that moves first is defined as a "Stackelberg leader" (Sinclair and Stabler 1997:115­ 119). In each period, the local government obtains tax receipts given by the tax multiplied by the size of the expanded capacity. The second part of the government's objective function consists of the net value derivable from the use of the land. Such a value is the net outcome of the government's evaluation of the public benefits from development against its environmental costs. On the one hand, public benefits are generated by tourism activities through a multiplier effect on the local economy arising because a local workforce is used in con- struction, in operating the tourism facilities and because tourists con- sume local products (Wunder 2000). On the other hand, land use engenders environmental costs for the local population both in terms of congestion and exhaustion of the natural resource and in the form of loss of non-market benefits such as bequest, option and existence values (Pearce et al 1989). The bequest value is the willingness to pay to preserve the environment for the benefit of future generations. The option value identifies an expression of preference for the preservation of an environment against some probability that a community will make use of it at a later date. The existence value is represented by the utility that individuals enjoy when the risk to an endangered spec- ies has been reduced. Development implies the loss of all these values for the local population, the sum of which constitutes the non-market environmental costs. These are particularly high in areas where the ecosystem is less resilient (less likely to fully recover from exogenous shocks) and where, therefore, development may engender irreversible degradation to the environment (Barbier and Markandya 1990). Finally, land use engenders another form of environmental costs to the local population (the lack of access to natural resources, such as beaches) due to congestion arising from having an intensively developed destination occupied by tourists. The foregoing analysis of the local government's pay-off function 894 SUSTAINABILITY AND LAND TAXATION has highlighted that the development engenders benefits in the form of private profits and public income multiplier effects, and environ- mental costs whose size depends on the characteristics of the area's ecosystem. The difference between these benefits and costs gives rise to a net value function, W(B(t)), that depends on the size of the project. A realistic analytical expression for the net value function W(B(t)) is one that allows, at an early stage of development, growth benefits, and the related efficiency gains, to be greater than environmental costs, which become predominant as the project increases in size. A quad- ratic function such as W(t) = B(t) B2 (t), exhibits this property: the minus sign attached to the quadratic value of the total establishment's size, denoted by B(t), implies that the environmental costs associated with development, increase at a faster rate than growth benefits, cap- tured by the linear part of B(t). Furthermore, this effect is reinforced depending on the value of , that is, the larger , the larger the environ- mental costs relative to the growth benefits. However, when is suf- ficiently large (such as when the loss of existence, option and bequest values is conspicuous) W(t) may be negative even for small levels of tourism activity. The sum of the tax receipts and the net value function give the total pay-off for the local government: E(t) = (t)(t) + W(t) = (t)(t) + B(t) - B2 (t). Such an expression represents the total value that the local government attaches to the development. This is made up of the pecuniary tax take and of the external effects of the project, namely the growth benefits and the environmental costs. When is large enough, environmental costs exceed the monetary value of tax receipts and income multiplier effects even for small values of B(t) that correspond to an early stage of development. In this case, given that E(t) < 0, the local government should not allow any activity in the territory. How- ever, in more general cases, the tax receipts may be used to improve the site's infrastructures (schools, hospitals, roads, etc.) whose creation thus compensates future generations for inheriting a smaller stock of natural resource. This discussion highlights the different objectives pursued by the developer and the local government, because the taxation policy cre- ates a benefit for the latter and a cost for the former. However, without the tax, the external (that is, income multiplier and non-market environmental) effects that the tourism project engenders do not affect the developer's pay-off. Thus, the tax constitutes an instrument that induces the internalization of these external effects in the devel- oper's decision concerning the amount of land to develop. Moreover, it was previously noted that the local government can use taxation to appropriate some of the private profits. This guarantees the partici- pation of the local population in the sharing of the tourism develop- ment's benefits and, hence, intragenerational equity. It has also been argued that developments are unable to transfer an intact stock of natural resources to future generations. It follows that the analysis of intergenerational equity, and the related issues of conservation and efficiency, should be carried out by studying how the land tax 895CLAUDIO PIGA presented above should be set depending on the environmental characteristics of the territory. Results The following results are derived assuming that both the developer and the government choose, respectively, the size of newly developed land ((t)) and the land tax ((t)) in each period so as to maximize the discounted flow of profits over an infinite time horizon. The opti- mal tax is such that the local government's inter-temporal pay-off E(t) is maximized. The technical analysis used to derive the solution to this maximization problem is beyond the scope of this article. The inter- ested reader is referred to Piga (1999, 2003). To analyze how the tax affects the equilibrium value of land use, it is customary to derive the socially optimal result and use it as a bench- mark against which the model's results are compared. The best out- come, from a social viewpoint, is one in which both parties mutually agree on the size of the territory that is developed in every period. Such a decision entails that all types of benefits and costs identified above are taken into account. Hence, there is no need for the tax, as both parties understand the impact of tourism activity on the area, to the extent that no establishment is built if the environmental costs are greater than the private and public benefits. If development is under- taken, then a lump-sum transfer from the developer to the local government is sufficient to obtain intragenerational equity. Such a benchmark case can be denoted as the "cooperative case", to emphas- ize the fact that the parties behave as a single entity: this implies that all the external effects associated with the project are internalized and properly accounted for in the decision regarding the project's size. In the remainder of the paper, the equilibrium level of land use in the perfectly cooperative scenario is denoted as B (t). Such a case is equiv- alent to the partnerships model advocated in Middleton (1998:128­ 9), where a collaborative process is established between the private and the public institutions for dealing with tourism planning and manage- ment for a destination. However, it is very unlikely that the developer and the government have perfectly aligned objectives. While the cooperative case can be used as a benchmark, in a more realistic case the government and the developer behave non-cooperatively as two distinct entities pursuing conflicting goals. Indeed, it is reasonable to expect that the developer wants to maximize the project's private benefits without considering the social costs that it entails, while the government may be particularly concerned with its social costs and the loss of environmental values. From a strategic viewpoint, the government can exploit its "first-mover advantage" by imposing a development tax, (t), in order to induce the optimal level of land use which is equivalent to that in the cooperat- ive case. Indeed, if the developer were left free to operate without any form of public intervention, the establishment size would exceed that in the cooperative case, as the environmental costs would not be taken into account. However, even in the non-cooperative scenario, 896 SUSTAINABILITY AND LAND TAXATION obtaining the socially optimal level of land use is possible when the government behaves like a Stackelberg leader. Another advantage of the non-cooperative scenario is that it allows analyzing the effects that different weights used by the parties to dis- count future benefits and costs have on the resource use. Such weights are represented by the discount factors that capture the extent to which different individuals presently value a gain or a loss that will occur in the future. Those who prefer to make a lower gain today rather than a higher gain tomorrow have a greater discount factor. The length of the planning horizon over which investments are evalu- ated also influences the discount factor. Traditionally, private devel- opers are more concerned about the short-term implications of their strategies, and thus discount the future more heavily. When evaluating the impacts of human activity on the environment, governments are, at least in theory, more likely to consider the related long-term costs. Hence, more concern by the government for the intergenerational effects of tourism activity are represented by a lower discount rate. The results from the theoretical model reported below are obtained assuming that the developer's discount rate, denoted by D, is greater than the government's rate, denoted by G. However, the results are reversed if the local government discounts the future more than the developer, a situation that is more likely to occur in a certain phase of the political cycle, namely before an election. Having defined the model's components, the qualitative features characterizing the model's solution follow. First, a result common to the cooperative and the non-cooperative setting is that the largest por- tion of space is developed at the beginning of the development. The rationale behind such a choice is clear: it is optimal to build as much capacity as possible at the beginning of a project because an extra unit of capacity generates revenues forever. Thereby, the developer can more quickly recoup the large initial capital investment used to finance the high start-up costs associated with the project. Second, and more importantly, the total amount of used land in the non-cooperative case, denoted as Bnc (t,D,G), is the same as in the cooperative case, B (t), when the two parties have the same discount factor: that is, Bnc (t,D,G) = B (t) when D = G. It would seem that the two settings generate the same outcome with regard to the conser- vation issue. However, in the non-cooperative setting the two insti- tutions usually value future benefits and costs differently. This has pro- found bearings on the establishment size in equilibrium. Indeed, as assumed before, if the private developer is less patient, then the model clearly shows that in the non-cooperative case less land will be used in total relative to the benchmark. This corresponds to an equilibrium where Bnc (t,D, G) < B (t) when D > G. The rationale of such a result is intuitive. A government that places a great weight on the environmental costs that future generations will incur, can set the land tax high enough to deter development beyond Bnc (t,D, G). It is note- worthy that intergenerational concerns may not conflict with efficiency considerations, because a territory of size Bnc (t,D, G) is actually developed. 897CLAUDIO PIGA Third, it can be shown that the rate of exploitation of the natural resource occurs at a slower pace in the non-cooperative case, meaning less land (a smaller (t)) is used in every period. It follows that less environmental costs are also incurred. The results that in the non- cooperative case less land is used in total and that the development occurs at a slower pace, cast some doubts on the recommendation that the private and the public sector should seek more collaborative forms of organization of the tourism activity. While it can be argued that the same outcome could be achieved without the tax by having the government setting the limits of development (Bnc (t,D,G) and (t) in every t), and the developer agreeing to comply with these limits, such a command-and-control arrangement presents, relative to the market instrument of taxation, at least two drawbacks: it carries high bureau- cratic costs for monitoring and enforcement, and, most importantly, it may be easier for the parties to renegotiate the terms of the initial agreement because of changes in socioeconomic and/or political con- ditions, while changes in the tax legislation have to undergo a lengthy parliamentary scrutiny. Thus, a better outcome is obtained when the local government regulates the activity of the private developer by imposing a land tax, relative to a situation where a mutual agreement between the parties is created. Fourth, the model indicates how public policy adjusts according to the relative size of environmental costs and public benefits in the cost- benefit analysis. Foremost, regardless of whether the environmental costs are greater than the benefits from development or not, it is always optimal for the local government to impose a positive tax on the use of the natural resource at the outset of planning. Obviously, the tax remains positive whenever the environmental costs are greater than the public benefits due to endogenous growth and income multiplier effects. However, when the public benefits are larger than the environ- mental costs for any establishment size, the optimal tax's time profile exhibits an interesting behavior: at the outset of planning, the tax is positive but at some point in time it is turned into a subsidy, which corresponds to a negative tax. Traditional economic literature suggests the use of a subsidy to attract investments that determine beneficial effects for the host community, including jobs creation, higher local firms' birth rate, and more (Sinclair and Stabler 1997:200­210). An important difference in the present case is that it is optimal for the policymaker to introduce the subsidy only after an appropriate time period, and not at the outset. The rationale can be found in the ability of the tax to extract some of private benefits engendered by the project which otherwise would be totally appropriated by the firm. Finally, the analysis implicitly allows geographic areas to be taxed differently depending on the sensitiveness of the natural environment. Low environmental costs are associated with non-sensitive areas. The government initially imposes a low tax for development in such areas, which may then become a subsidy if public benefits are greater than environmental negative effects. On the other hand, when concerns for the resilience of the natural environment exist, the tax is set at such a high level that no project is economically profitable for the devel- 898 SUSTAINABILITY AND LAND TAXATION oper. This is tantamount to one of the four sustainable tourism approaches proposed by Hunter, that of "Neotenous Tourism", which corresponds to a "very strong sustainability approach ... predicated upon the belief that there are circumstances in which tourism should be actively and continuously discouraged on ecological grounds". Fur- thermore, Hunter argues that in "some places, including national reserves of national or international importance, tourism growth should be sacrificed for the greater good" (1997:852­863). Such a rec- ommendation is equivalent to the case of no development due to the high tax that arises as an equilibrium outcome in the model. To summarize, the results from the economic model of taxation show that it is possible to pursue conservation objectives without hin- dering the implementation of viable economic projects that impact the environment. The results crucially hinge on the assumption that the government has a concern for conservation, which in practice is ident- ified by its willingness to give proper weight to environmental damages that may occur in the distant future. As argued at the outset of this paper, such willingness can be expressed by the creation of a legislation that has constitutional value. The following case study illustrates the crucial role that regional planning legislation plays for the establish- ment of a government's stance towards conservation. The Case of Costa Smeralda in Sardinia The previous analysis is based on the properties of a normative model that encompasses a set of economic issues. A test of its predic- tions is a case study involving a local government and a tourist devel- oper. The evidence presented suggests that the recommendations derivable from the model's results provide a set of principles that local governments apply--or should begin to apply--in their evaluation of tourist projects. In 1997 a multinational company, Ciga Immobiliare, prepared a devel- opment scheme for a part of Costa Smeralda on the Northeast coast of Sardinia in the Mediterranean Sea. The project was literally called "masterplan" in the Sardinian and national press, and this is how it will be referred to in the remainder of this article. The following fig- ures are taken from the executive summary released by the developer when the project was presented to the press. The entire project was to cover an area of 24 km2 , 85% located in the territory of the Arzachena municipality and the remaining 15% in the Olbia municipality terri- tory. The total volume of the 11 hotels, 2,000 villas, and 1,900 apart- ments included in the project was estimated to be 2,550,000 m3 (14% in hotels, 9% in ancillary services, and 77% in residential dwellings). The planned capacity of hotels and residential dwellings was, respect- ively, 4,000 and 16,000 beds. The cost of the total investment was calcu- lated to be $1.45 billion and the related increase in local income to be $2.4 billion, with an investment multiplier of 1.7. Although the pro- ject was to cover a time span of 25 years, most construction would be completed within the first 5­10 years. The yearly demand, measured in bednights, was evaluated at 504,000 899CLAUDIO PIGA units in hotels and 960,000 units in residential dwellings, the total being almost three times higher than the average of 560,000 units recorded during the 90s in the already existing hotels owned by the developer in Costa Smeralda. The expenditure by tourists in hotel accommodation and dwellings was expected to total, respectively, $151 and $240 millions and generate an increase in local production of around $500 millions. Further, the developer's study estimated the overall increment in employment in Sardinia to be 12,200 jobs: 4,800 new jobs in hotels and restaurants, 2,000 in trade and retailing, 4,200 in the services, and 1,200 in manufacturing. Given these figures, the project should certainly have appeared enticing to a local government operating in a region characterized by an extremely high unemployment rate and a much lower per capita income than the Italian average. How was it, then, that the parties did not find a satisfactory agreement and eventually the developer with- drew the project? An answer can be found by analyzing the political stance of the Sardinian government with respect to the magnitude of the social and environmental costs associated with the project. The Sardinian government had made it clear that the "safeguard of the environmental quality is a factor for economic development" (Sardinian Regional Government 1997:7). Moreover, as far as the mas- terplan was concerned, the position of the Sardinian government was that "a better balance between the use of the territory, the socioecon- omic effects and the developer's profit has to be found". The Sardinian government seemed inclined to grant the building permission, subject to a few changes in the project, including a significant reduction in the number of residential dwellings, and possibly an increase in the number of hotels; priority to the building of hotels and tourism facili- ties; and respect for the norm that prohibits the construction of any building within 300 meters of the coastline. The local government expressed concern about the number of "scattered dwellings"--pre- sumably villas--which "consume land and make access to the sea more difficult". The Sardinian government's preference for hotels and the request to increase the number of hotels can be found in the regional government's belief that hotels and related facilities would extend the tourism season well beyond the peak months of July and August. The Sardinian government's remark that villas and apartments use more land than hotels is consistent with the theoretical approach and indi- cate the government's awareness of the environmental costs associated with the project. The requirement to significantly reduce the number of residential dwellings is directly linked to the theoretical analysis (Tables 1 and 2). The data in Table 1 include census statistics collected in 1981 and 1991. Rows (1), (2), (3) and (4) indicate that the number of dwellings increased sharply, thereby suggesting that intensive building activity took place during the decade in the municipalities of Olbia and Arza- chena. A similar phenomenon occurred also in the remaining part of the North Sardinian Province, which includes many other seaside resorts. A first consideration is the obvious positive correlation between the change in the number of resident population (row 11) and the 900 SUSTAINABILITY AND LAND TAXATION Table 1. Census Statistics in North Sardinia Arzachena Olbia Province Variables 1981 1991 1981 1991 1981 1991 (1) Occupied dwellings 2349 3155 8295 13811 121186 145114 (2) Unoccupied dwellings 4690 7379 6954 10528 46789 71024 (3) Of which: usable for 3004 6394 4187 8190 25302 47119 holidays (4) Total dwellings (1)+(2) 7039 10534 15249 24339 167975 216138 (5) Ratio (3)/(1) 1.279 2.027 0.505 0.593 0.209 0.325 (6) Rooms in occupied 10182 13987 36736 59994 535798 666602 dwellings (7) Rooms in unoccupied 15767 26292 20752 33990 156306 231132 dwellings (8) Of which: usable for 10374 23238 12906 25745 87998 151278 holidays (9) Avg. rooms in occ. dwelling 4.33 4.43 4.43 4.34 4.42 4.59 (6)/(1) (10) Avg. rooms in holiday dwell. 3.45 3.63 3.08 3.14 3.48 3.21 (8)/(3) (11) Resident population 7998 9435 30787 41095 433842 454904 (12) Size of territory (Km2 ) 228.59 228.61 388.73 376.10 7519.91 7519.93 (13) Dwelling density (4)/(12) 30.79 46.08 39.22 64.71 22.33 28.74 Source: ISTAT, Istituto Nazionale di Statistica, Censimenti della Popolazione, Years 1981 and 1991. increase in the number of occupied dwellings. However, more relevant to the present analysis is the drastic increase in the number of holiday dwellings, whose number more than doubled in Arzachena and almost doubled in Olbia. Row (5) reports the ratio of holiday dwellings over occupied dwell- ings. It shows that, relative to the rest of the province, the two munici- palities are characterized to a great extent by the so-called "second homes" phenomenon (holiday houses left unoccupied for most of the year) (Jaakson 1986). In terms of the theoretical analysis, the extensive use of land associated with this type of accommodation entails high environmental costs, deriving for example, from loss of flora and fauna, beach erosion, and degradation of water quality. Other environ- mental problems relate to the fact that second homes are used mainly during the period of peak demand. Moreover, self-catering accommo- dations, such as second homes, do not generate as high multiplier effects as hotels. The masterplan might have exacerbated the over- crowding and congestion. Indeed, the dwelling density for the mas- terplan--3900/24=162.5 dwellings per Km2 (3900 is the sum of 2000 villas plus 1900 apartments, 24 Km2 is the size of the territory, above)-- is well above the averages of the territories concerned (row (13) of Table 1). This is indicative of how intensively some parts of the space would have been used by the developer. As many of the masterplan's 901CLAUDIO PIGA Table 2. Comparison of Total Building Activity by Location Total Building Activity Arzachena Olbia Masterplana (Sum 1982­1996) (1) Volume of new residential buildings 788505 834450 1,963,500 (m3 ) (2) Volume of new non-residential 260854 1,113,830 586500 buildings (m3 ) (3) Total new volume (m3 ) ((1)+(2)) 1,049,359 1,948,280 2,550,000 (4) Number of new residential dwellings 2630 2845 3900 (5) Number of new rooms in new 8787 11332 11700b residential dwellings Source: Istat, Istituto Nazionale di Statistica, Statistiche sull'Attivita' Edilizia, Years 1982--1996. a Source indicated in the main text. b Estimated considering an average of 3 rooms per dwelling. buildings would have occupied areas characterized by high environ- mental sensitivity, it can be inferred that the social costs deriving from the loss of the "hidden environmental values" would have been con- spicuous. Table 2 provides a summary of the building activity undertaken in the two territories over the period 1982­1996. The variables in this table are the volumes built for residential and non-residential pur- poses, the number of new residential dwellings and the related number of rooms. They are reported as the sum over the period 1982­1996. These values are compared in the table with those corresponding to the masterplan's blueprint. Volumes are used in the following analysis for various reasons. In Italy planning regulations dictate the maximum allowable volume of a building erected in a territory of given size. Stat- istics are collected accordingly. Further, volume is a good indicator of building activity because territorial planning imposes restrictions on the total height of buildings. Therefore, volumes are highly correlated with the size of the territory occupied by buildings. The volume of residential buildings completed over the period in the two municipalities is less than the amount planned by the mas- terplan's developer (row 1). The comparison between the figures in Table 2 and the size of the project is made even more striking by the consideration that only 15% of the project was planned in the Olbia municipality. This implies that 85% of the residential volumes intended in the masterplan (totaling 1,668,975 m3 ) would have been built in the Arzachena's territory. This exceeds the total volumes built there for residential and non-residential purposes over the period 1982­1996 (row (3)). Moreover, an inspection of rows (4) and (5) reveals that the number of residential dwellings and rooms intended in the masterplan is greater than the number of residential dwellings and rooms actually built in each single municipality over the period 1982­1996. 902 SUSTAINABILITY AND LAND TAXATION Therefore, Table 2 provides support for the notion that the total size of the masterplan was at least equivalent to, if not greater than, the overall residential building activity in the two municipalities over the period 1982­1996. Hence, had the project been allowed to pro- gress as originally planned, many future development options in the area would have been foreclosed, thereby engendering further social costs in territories where building activity for tourism has been parti- cularly intensive over the last two decades. The rationale for the aforementioned condition (respect for the norm that prohibits construction within 300 meters of the coastline) lends support to the choice in the theoretical model of considering the local government as endowed with a first-mover advantage enabling them to pursue long-term conservationist policies. Indeed, the rationale of such a norm is that of clearly declaring the Sardinian government's commitment toward the safeguarding of the coastal environmental quality. The exception granted to the masterplan's developer could easily have become the rule, and led to future greater environmental costs from over-exploitation. When it became clear that the project had to be changed substantially in its residential compo- nent and that the local government would not have given any per- mission to build within 300 meters of the coastline, the developer decided not to exercise the option to buy the territories from the cur- rent owner (the American multinational Starwood) and abandoned the project. The analysis of this case study highlights the following points that are consistent with the theoretical predictions from the model discussed in the previous section. One, the concern of a local government for the preservation of environmental quality may lead to no development. Two, the local government's behavior emphasizes the importance of maintaining existing legislation aimed at conservation, thereby giving it the status of constitutional dictate. Three, and more generally, the study shows how the variables considered in the theoretical analysis are normally taken into account in the blueprint of any investment project in the tourism industry. CONCLUSION This study argues that a limitation of the strong sustainability approach is the lack of emphasis given to efficiency considerations. It suggests the application of the two-tier approach outlined in Page (1997) as a conceptual framework to tackle the issues of intergener- ational equity and economic efficiency simultaneously. Such a two-tier method conceives the importance of the natural resource, the identi- fication of the appropriate instruments and the legislative framework as crucial elements. To provide further support to Page's recommen- dations for the case of tourism, this study has presented a theoretical economic model of taxation and a case study involving a local govern- ment and a private developer. When economic efficiency considerations are paramount, the theor- etical model suggests the use of a land tax at the beginning of the 903CLAUDIO PIGA development followed by the possible introduction of a subsidy at a later stage. The tax guarantees that local communities can extract some of the rents created by tourism activity that otherwise would entirely accrue to the developer. At the other extreme, when the devel- opment endangers the existence of an essential resource or ecosystem, the theoretical model suggests the prevention of any form of develop- ment. In intermediate cases, the tax on land use combines efficiency and conservation goals, as the tax curtails, but does not prevent, tour- ism activity. This situation is represented in the case study where emphasis is given to the crucial role of territorial planning legislation for the reduction of negative impacts of tourism. Indeed, in the two municipalities of Arzachena and Olbia the number of holiday homes has grown dramatically over the last two decades, thus suggesting the economic viability of such developments. At the same time, the norm prohibiting constructions within 300 meters of the coastline has prevented the irreversible damage that building on the coast entails, and has forced developers to locate their sites in less sensitive areas. The fact that the Sardinian government chose not to create a precedent by allowing special treatment for the masterplan's developer, suggests that the norm regulates issues that deserve to be treated within a constitutional framework, in line with Page's recommendations. Unfortunately, the norm above does not have a legal constitutional status, and could thus be changed using the process applicable to any ordinary law. This is indeed the intention of the new Sardinian government elected in the year 2000 (to modify legislation on territorial planning to accommodate the possibility of buildings near the coastline). Both the model of taxation and the Sardinian case study provide support for the argument in Middleton (1998) that governments do not need to introduce new policy instruments, but rather, they should gear the existing ones towards the achievement of sustainable objec- tives. Thus, territorial planning, building regulation, provision of infra- structure, fiscal incentives and disincentives, ecological labeling (Mihalic 1996), assessment and management of carrying capacity (Collins 1999), and information and education of tourists (Filho 1996) can all be used effectively and play a central role in a public strategy for sustainable tourism. However, the view advocated by Middleton (1998) that the public and the private sector should seek more integrated forms of organiza- tion in the management of tourism does not necessarily follow. The theoretical analysis of taxation clearly indicates that the best outcome, both in terms of conservation and overall efficiency, is obtained when the public sector takes advantage of its ability to set the rules of the game. The Sardinia masterplan example shows that negotiations with the private sector may turn out to be unbeneficial in the long run, especially if the private developer requests special exemptions from the existing legislation, which is aimed at limiting the environmental impact of tourism projects. Whenever the government and the private developer have conflicting views regarding the development of an environmentally sensitive area, the analysis of the theoretical model, 904 SUSTAINABILITY AND LAND TAXATION and the support it has received from the case study, suggest that the government should exploit its first-mover advantage, which is incom- patible with Middleton's bargaining approach. As a corollary to the analysis, it is concluded that private-public partnership may be favored whenever intergenerational equity issues are not a crucial concern. A Acknowledgements--The author thanks Enrico Bachis, Aoife Hanley, and Thea Sinclair for their help and advice. REFERENCES Barbier, E., and A. Markandya 1990 The Conditions for Achieving Environmentally Sustainable Develop- ment. European Economic Review 34:659­669. Collins, A. 1999 Tourism Development and Natural Capital. Annals of Tourism Research 26:98­109. Cronin, L. 1990 A Strategy for Tourism and Sustainable Development. World Leisure and Recreation 32:12­18. Faucheux, S., D. Pearce, and J. Proops, eds. 1996 Models of Sustainable Development Cheltenham: Edward Elgar. Filho, W. 1996 Putting Principles into Practice: Sustainable Tourism in Small Island States. In Sustainable Tourism in Islands and Small States: Issues and Policies, L. Briguglio, B. Archer, J. Jafari and G. Wall, eds., pp. 61­68. New York: Pinter. Font, A. 2000 Mass Tourism and the Demand for Protected Natural Areas: A Travel Cost Approach. Journal of Environmental Economics and Management 39:97­116. Gunn, C. 1994 Environmental Design and Land Use. In Travel, Tourism, and Hospitality Research. A Handbook for Managers and Researchers, B. Ritchie and C. Goeldner, eds., (2nd ed.), pp. 243­258. New York: Wiley. Howarth, R. 1997 Sustainability as Opportunity. Land Economics 73:569­579. Hunter, C. 1997 Sustainable Tourism as an Adaptive Paradigm. Annals of Tourism Research 24:850­867. Hunter, C., and H. Green 1995 Tourism and Environment. A Sustainable Relationship? London: Rout- ledge. Huybers, T., and J. Bennett 2000 Impact of the Environment on Holiday Destination Choices of Prospec- tive UK Tourists: implications for Tropical North Queensland. Tourism Eco- nomics 6:21­46. Jaakson, R. 1986 Second-Home Domestic Tourism. Annals of Tourism Research 13:367­ 391. Keane, M. 1997 Quality and Pricing in Tourism Destinations. Annals of Tourism Research 24:117­130. Middleton, V. 1998 Sustainable Tourism: A Marketing Perspective. Oxford: Butterworth-Hei- nemann. Mihalic, T. 1996 Ecological Labelling in Tourism. In Sustainable Tourism in Islands and Small States. Issues and Policies, L. Briguglio, B. Archer, J. Jafari and G. Wall, eds., pp. 197­205. New York: Pinter. 905CLAUDIO PIGA Norton, B., and M. Toman 1997 Sustainability: Ecological and Economic Perspectives. Land Economics 73:553­568. Page, T. 1977 Conservation and Economic Efficiency. Baltimore: John Hopkins Univer- sity Press. 1997 On the Problem of Achieving Efficiency and Equity, Intergenerationally. Land Economics 73:580­596. Pearce, D., A. Markandya, and E. Barbier 1989 Blueprint for a Green Economy. London: Earthscan Publications. Piga, C. 1999 Pigouvian Taxation and Sustainable Development in Tourism. Christel DeHaan Tourism and Travel Research Institute Working Paper series, 99/1, Nottingham University Business School: 2003 Pigouvian Taxation in Tourism. Environmental and Resource Econom- ics, forthcoming. Sardinian Regional Government 1997 Audizione della Giunta Regionale sul "Master Plan": Seduta congiunta delle Commissioni permanenti Quarta e Sesta del Consiglio Regionale della Sardegna. Report of the Regional Government on the "Master Plan" to the Fourth and Sixth Permanent Committees of the Sardinian Regional Parlia- ment, pp. 1­14. Cagliari, October. Sinclair, T., and M. Stabler 1997 The Economics of Tourism. London: Routledge. Toman, M. 1994 Economics and "Sustainability": Balancing Trade-off and Imperatives. Land Economics 70:399­413. WCED 1987 Our Common Future. Oxford: Oxford University Press. Wunder, S. 2000 Ecotourism and Economic Incentives: An Empirical Approach. Ecologi- cal Economics 32:465­479. Submitted 7 September 2001. Resubmitted 8 May 2002. Resubmitted 19 February 2003. Accepted 20 March 2003. Final version 17 April 2003. Refereed anonymously. Coordinating Editor: Lisle S. Mitchell