بررسی مالیات بر سوخت های فصلی در یک مقصد گردشگری انبوه : مطالعه موردی برای جزایر بالئارس

This paper estimates the monthly aggregate demand for diesel oil and gasoline in a mass tourism region, characterized for a high level of seasonality. Using time series models, price elasticities are estimated with special emphasis in evaluating differences between seasons in order to assess the consequences of a fuel tax applied exclusively during the high season. Using the case study of the Balearic Islands (Spain) from January-1999 to December-2010 results from a partial adjustment model show a relatively low price-elasticity, evidencing how the internalizing mechanism that could be argued for introducing the tax in order to reduce transport externalities does not work. Additionally no statistical differences have been found between seasons for both fuels invalidating the argument that tourism activity reacts differently to host activity.

Petroleum products demand has received a great deal of attention as a research topic during the last decades. Initially, since the 1973 oil crisis, a growing number of studies modeled demand for gasoline addressing concerns about the availability of this non-renewable resource. Most of those studies focus on the demand for motor gasoline for automobiles since the segment was relatively important and represented one of the highest growth rates. The quantification of price and income long-run elasticities of fuel demand was of paramount interest in order to project future trends of oil markets and to plan infrastructures and strategic reserves (Dahl and Sterner, 1991). Whereas the last late century scientific research was fueled mainly by the threat of energy scarcity, nowadays environmental problems like the potential global warming change are becoming increasingly important. Recently, studies have directed the interest to the various environmental consequences of petrol consumption, particularly with respect to the emission of greenhouse gases (Gallo, 2011, Kyle and Kim, 2011 and Marrero, 2010). In this new context, accurate estimations of petrol demand are also important because of the wide range of fiscal instruments that are worldwide applied. The need for mechanisms that promote more efficient use of transport has encouraged the use of fuel taxes or charges, given their considerable benefits in terms of the achievement of a “double dividend” through the improved efficiency of the tax system (Bovenberg and Mooijr, 1994) and efficient compliance with the “Polluter Pays Principle”. Because tourism is one of the most transport-intensive sectors, these principles have led to different purposes aimed to directly tax those activities less environmentally respectful, particularly the increase of fuel prices which is one of the last options that have recently emerged in the context of tourism policy (Mayor and Tol, 2010, Rothengatter, 2010, Tol, 2007 and Zhang et al., 2010). Recently Aguiló et al. (2012) contemplated the possibility of taxing gasoline and diesel oil fuels only during high season months in those destinations characterized by a high level of seasonality. The theoretical arguments for a temporal discriminatory fuel tax during the year include the high exportability of the tax, especially if it is applied during the high season an important part of the revenues from taxes will come from non-residents (e.g., by non-voters), and discouraging private road transport during the high season, which is one of the most important tourist externalities (Palmer et al., 2007). The objective of this paper is to model and estimate diesel oil and gasoline demand with the aim of assessing the application of a fuel tax applied exclusively during the high season months. The paper is structured as follows. In Section 2 the relationship between gasoline demand and their determinants is discussed, with reference to the seasonal contribution of the tourism sector. In Section 3 methodological considerations and model specification are introduced. Section 4 presents data and results. Section 5 discusses policy implications and concludes.

In this study diesel oil and gasoline demands have been estimated in the context of a tourist region characterized by high seasonal fluctuations of population on the territory. The case study of the Balearic Islands (Spain) have been undertaken showing, as expected, that fuel consumption responds positively to changes in income and negatively to changes in prices but also to human pressure which is a significant factor in explaining diesel oil demand. In summary, diesel oil and gasoline seem to have the same average price elasticity which tends to unity. In the short-run, gasoline consumers have statistically the same sensitivity among the different seasons, whereas diesel consumer response seems to be inelastic in the high season. The short-run price elasticity measures the change of the driving behavior as a result of a change in the price of fuel. The driving response to an increase of prices could be materialized in a reduction of distance travelled, increase in the fuel efficiency of driving or shift to another transport mode (i.e., public transport). This implies that gasoline consumers will generally respond to price changes regardless of the season of the year. Diesel drivers appear to be more sensitive to price changes in the low season than in the high season. One hypothesis is that economic activity in Balearics, which is mainly touristic and has a significant seasonal aspect, has led to an increase in mobility within and between touristic regions (i.e., Palma, Arenal, Alcúdia…) during the high season. It is fair to mention that limited accessibility of public transport to some touristic places reduces the options of travel mode, hence increasing fuel consumption. Regardless of whether the consumer is a resident or tourist, tourism activity seems to be the main factor that drives mobility in Balearics and consumers have less ability to respond to price changes during the high season. In the long-run elasticities are expected to increase, with gasoline consumers being more responsive in low season than high season. Diesel consumer showed no significant variation among the different seasons of the year. In addition, human pressure exerts an important role in modeling fuel demand. For instance, in high season, tourist pressure is highly significant for gasoline model, whereas, the significance of resident pressure is limited to diesel demand. If we assume that mobility in the low season is mainly driven by residents more than tourists, one could conclude that residents tend in the long-run to shift to diesel cars. Given that the fuel efficiency substitution options are quasi null for diesel cars, one would expect that in the long-run response to diesel price changes would be insignificant among seasons for the year. Estimated elasticities can be used to analyze different price and tourist policies. Thus, from the tax analysis, the relatively low price-elasticity in the short-run shows how the internalizing mechanism that could be argued for introducing the tax in order to reduce transport externalities does not work. Although, an increase in fuel prices should be used as an effective tax collection instrument, that would benefit for a high level of exportability (taxes would be paid in a high rate by non-residents, that is, by non-voters) the homogeneity tests for price elasticities in the low and high seasons shows that applying these taxes regularly and only during the high season cannot be justified in differences in price elasticity. Results indicate that if an environmental surcharge is added to gasoline taxes, then the additional tax will decrease gasoline consumption only slightly and, therefore, will have little effect. For example, the estimated price elasticity estimates suggest that a hypothetical increase in gasoline and diesel prices of 0.05€ exclusively during the high season would decrease gasoline and diesel consumption in the short-run by 8.78 and 5.70 million liters, respectively (in percentage terms a reduction of 5.6% and 2.8%, respectively, during the high season). If the increase in gasoline and diesel prices was during the whole year the decrease in gasoline and diesel consumption in the short-run would be 21.17 and 30.85 million liters, respectively (in percentage terms a reduction of 5.2% and 6.1%, respectively, during the year). The numbers contrast the preconceived idea that fuel reduction should be relatively pronounced during the high season. The percentage changes, part of being small, show a slight difference between high and low seasons in the case of gasoline. For diesel, the results point out that, relatively, significant reductions are made in the low season, and therefore residents would be affected more. Thus, as far as this analysis is concerned, it is not apparent that it would be worth pursuing such an inappropriate tax for a small reduction in fuel demand which in turn produces little improvement in the environment. New trends in tourism point towards an increase in tourist mobility in the host region. Thus, findings of this study appear to be significant policy implications for specialized tourist economies, particularly with respect to the way in which taxation could help government increase fiscal revenues and regulate the level/structure of fuel consumption from a temporal point of view. In any case, as a limitation of this study, it should be noted how price elasticities have been estimated, considering both residents and tourist populations jointly. Although, results are valid in mean terms, future research will have to focus in trying to estimate if different price elasticities characterize different fuel consumer groups and to identify if income classes exhibit different price elasticities, an issue that would have distributional implications. An additional limitation of this study arises in the fact that a complex relationship does exist between tourism, income and fuel demand. Changes in GDP could be due to the changes in the tourism activity which is the main economic sector in a tourist area. Thus, the implicit assumption in this study, is that tourism has a direct effect on diesel oil demand, whereas the effect of other sectors is induced by the GDP variable in the specification model. The inclusion of the GDP should therefore be supported as a proxy of long term development that could be related to the intensity of the use of road transport and the efficiency of the vehicle stock.