Elasticities relevant to tourism in Scotland: evidence review

4 Behavioural changes as a result of tourist taxation

4.1 Introduction

This section reviews the existing literature on the impact of tourism taxes on tourism demand, including the number of tourist arrivals and tourist expenditure, in European contexts. A particular focus is placed on tourist tax, which is also called the hotel occupancy tax. Other tourism-related taxes such as tourism VAT and air passenger duty (APD, also called airport tax) are reviewed briefly too.

There has been very limited research on this topic. In total, five reports and four journal articles published since 2010 are identified. Among them, four reports (Aberdeen City Council, 2017; Grater London Authority, 2017; Scottish Government, 2019; All-Party Parliamentary Group for Hospitality, 2019) and one literature review article (Goktas and Polat, 2019) summarised existing evidence from the previous literature instead of carrying out primary research. Given the focus of the present report on primary research, the above reports are excluded, and this section will review the other identified literature. A summary of the primary research on the impacts of European tourism taxes is shown in Table 4.1.

Table 4.1 Summary of the primary research on European tourism taxes (2010-2020)

Organisation/Author: European Commission (2017)

Title: The impact of taxes on the competitiveness of European tourism

Type of literature: Report

Country/Region: EU

Tax type: Occupancy tax, air passenger duty, VAT, real estate tax, etc.

Organisation/Author: Callaghan and Tol (2013)

Title: UK tourists, the Great Recession and Irish tourism policy

Type of literature: Journal article

Country/Region: Ireland

Tax type: VAT, airport tax

Organisation/Author: Kristjánsdóttir (2020)

Title:Tax on tourism in Europe: Does higher value-added tax (VAT) impact tourism demand in Europe?

Type of literature: Journal article

Country/Region: Europe

Tax type: VAT

Organisation/Author: Seetaram, Song and Page (2014)

Title: Air passenger duty and outbound tourism demand from the United Kingdom

Type of literature: Journal article

Country/Region: UK

Tax type: Air passenger duty

According to the literature review by Goktas and Polat (2019), 19 EU countries have introduced a tourist tax. The tax rate varies between €0.10 (in Bulgaria) and €8.0 (Netherlands) per person per night, or between 1% (Romania) and 7% (Netherlands). The rate also varies across municipalities within the same country. Three types of levy have been implemented in Europe: the progressive visitor levy option, the fixed-rate or flat-fee visitor levy option, and the tax liability visitor levy option (The Scottish Parliament, 2018).

4.2 Review of Individual Studies

4.2.1 The impact of taxes on the competitiveness of European Tourism

This is a comprehensive study on the impact of tourism-related taxes in Europe. Twenty-seven EU states (including the UK) were included in the study. A range of tourism-related taxes such as air passenger duty, VAT, real estate tax and occupancy tax in EU countries were reviewed. In particular, the impact of the proposed occupancy tax rate change on tourism spending and service providers' revenue was estimated.

Firstly, this study presented some theoretical arguments on the impact of tourism taxation. By distorting price signals, taxation can lead to a reduction of economic efficiency and welfare. After the application of taxation, the price of a product is increased, and the quantity consumed and supplied is both reduced. As a result, both consumer welfare and producer welfare fall, part of which can be captured by the government in the form of tax revenues. The size of the net loss to society depends on both the absolute level of the tax imposed and the price elasticities of demand and supply. On the other hand, some academics recommended that tourism taxation could be used as a means of correcting negative externalities caused by tourists such as environmental repercussions (e.g., Chang, Lu and Hu, 2011).

This study further summarised empirical evidence from past literature, which generally suggested that the income elasticity and price elasticity of tourism demand are relatively high compared to other sectors. Therefore, a small change in taxation could potentially have a disproportionate impact on tourism levels if the reduction or increase in taxes is passed onto tourists in the form of lower or higher prices. The actual impact depends on the level of the price elasticity and the extent of pass-through to tourists. Between VAT and accommodation occupancy taxes, this study argues that the latter tend to have a more moderate effect on tourism demand, but they are likely to induce a psychological impact on tourists and could affect repeat tourism. However, the study finds inconclusive evidence from past literature on the overall effect of increased accommodation costs due to the occupancy tax. For instance, Bonham and Gangnes (1996) found that the impact of the charge levied in Hawaii on net hotel revenues was minimal, given inelastic demand for tourist accommodation. On the other hand, based on an econometric model supplemented with an input-output type of economic impact modelling, a study by American Economics Group (2004) suggested that a 2% increase in the combined tax on hotel and motel rooms could reduce room sales by 2.4%.

Similarly, this report showed inclusive evidence with respect to the impact of other tourism-related taxation. Using the Balearic Islands in Spain as a case study, this report revealed that, during the period between the announcement of the Eco-Tax in 2001 and its repeal, the Balearic Islands saw a 2-year decline to international tourist arrivals. The Eco-Tax was regarded to play a large role although other factors potentially contributed to this decline too. On the contrary, the Balearic Islands saw an 11.2% increase in international tourist arrivals in 2016 despite the introduction of the Sustainable Tourism Tax in 2015, and its official commencement on 1 July 2016. The statistics implied a minimal impact of this new tourism taxation.

This report pointed out that the existing research on the impact of tourism taxation was limited and less up to date. Therefore, further research is necessary to draw more robust conclusions.

Based on the above theoretical framework and available empirical evidence, this report further presents their study. With respect to the methodology, this study estimated the current tourism spending and service providers' revenue based on some assumptions of the demand structure (leisure vs business travellers) and the type of destinations (coastal vs non-coastal). Then different PED values from the previous literature were applied to different segments above. For the PED for business travellers, the report referred to the estimate of Peng et al. (2015)'s meta-analysis^[11] (-0.35); for leisure travellers, the report used the PED for Spain (-1.39) as a representative for coastal holidays and the PED for Germany (-0.43) for non-coastal holidays, both estimated by Smeral (1994). In addition, different from previous reports, different rates (60% and 100%) of the levy pass-through (to tourists) were assumed.

Regarding the impact of occupancy tax rate increase on tourism spending, this study found that the effect could be either positive or negative, depending on the relative shares of different tourist segments (business, coastal leisure and non-coastal leisure). In countries with a large share of coastal tourism such as Cyprus, Greece and Spain, an increase in occupancy taxes would lead to a decrease in tourism spending. On the other hand, for countries with a larger share of non-coastal or business tourism, an increase in taxes would increase total spending. In the case of these countries, the decrease in demand as a result of the higher price is not enough to offset the impact of the higher price paid per unit. Therefore, the impact on total spending is positive, despite the fall in demand. In the case of the UK, a new 3% occupancy tax would lead to a 0.32% increase in tourism spending but a 1.37% decrease in hotel revenues, assuming a pass-through rate of 60%.

4.2.2 Callaghan and Tol (2013)

Callaghan and Tol (2013) used a pooled travel cost model for UK travellers to estimate price elasticities of tourism demand and estimated the impacts of the proposed VAT and airport tax reductions in Ireland on visitor numbers and expenditure, as well as tax revenue. Given the average estimated PED value between -1 and 0 (i.e. price -inelastic), the effect of the abolition of the €3 airport tax would lead to a moderate tourism demand increase of 0.92%, and a reduction of the VAT rate (from 13.5 per cent to 9.0 per cent) would lead to a moderate tourism demand increase of 1.17%; in comparison, the increase in expenditure was small compared to the foregone tax revenue. The authors then concluded that "the Irish government is wrong to seek to stimulate inbound tourism through tax breaks. While effective, the costs exceed the benefits. It would be better, in fact, to increase travel taxes and VAT for tourism products. This would harm visitors and the tourism industry, but it would benefit the general population in the form of an increased tax take" (Callaghan and Tol, 2013, pp.113-114).

4.2.3 Seetaram, Song and Page (2014)

Seetaram, Song and Page (2014) estimated the impact of air passenger duty (APD) on UK outbound tourism demand. Based on the estimated ADLMs, both PED and demand elasticities of tax were calculated. According to the calculated PED values, UK tourists were found to be more responsive to changes in tourism prices at short-haul destinations than at long-haul destinations (-0.89 versus -0.37). With regard to tax elasticities, UK's outbound tourism demand did not appear to be responsive to tax changes regardless of long-haul destinations (-0.78) or short-haul destinations (-0.17) being concerned. Therefore, this study concluded that the effectiveness of APD was marginal in terms of reducing travel and the associated environmental impact.

It should be noted that the inclusion of both the tourism price and tourist taxation variables into the econometric models may affect the magnitude of the estimated elasticities due to the possible problem of multicollinearity in the estimated models.

Although this was acknowledged in the article, the authors did not carry out any statistical test on the existence of multicollinearity or estimate the potential effect on the magnitude of the estimated elasticities.

4.2.4 Kristjánsdóttir (2020)

Kristjánsdóttir (2020) estimated the relationship between VAT rates and tourist flows in Europe. Based on a basic regression analysis, the author did not find any negative relationship, but the author did not report the elasticities. Given the model specification and the available information presented in the article, it is not possible to calculate the elasticities. It should be noted that the regression model omitted key economic variables such as tourism prices and tourist income, and therefore the estimated effect of VAT was likely to be biased. This is evidenced by the extremely low model fits (R²< 0.04).

4.3 Summary

This section reviews empirical evidence in the literature on the behavioural impact of tourism taxation, with a particular focus on primary research. There is limited academic and grey literature available. One European Commission report and three academic articles were identified and reviewed. A general understanding is that an increase (or decrease) in tourism taxes tends to lead to a decrease (or increase) in the quantity of tourism demand, with other factors held unchanged. Considering the price effect of taxation, this is in line with the law of demand in economic theory, which reveals the inverse relationship between the price of a product and the quantity demanded. However, the impact of an increase in tourism taxation on tourists' expenditure would depend on the tax in question, consumers' PED, the PES, and other factors influencing both demand and supply. For example, different groups of tourists face different price elasticities. In general, business travellers and non-coastal holidaymakers tend to have a price-inelastic demand, while leisure travellers particularly coastal holidaymakers are likely to have a price-elastic demand. Therefore, an increase in tourism taxation would likely result in more tourism spending by business travellers and non-coastal holidaymakers, but less spending by leisure travellers especially coastal holidaymakers. It should be noted that most of the studies on the impact of tourism taxation focused on the price effect of tourism taxes. Tourist behaviour is affected by many non-price factors as well, such as advertising and news. More comprehensive research is necessary, which takes account of these multiple effects, in order to reach a fuller understanding of the overall effect of tourism taxation on behavioural changes.