Endogenous amenities, tourists’ happiness and competitiveness

ABSTRACT

A key strategy for supporting destination competitiveness is to enhance endogenous amenities, and tourists are the best candidate to evaluate them at the destination. The analysis in this paper uses a comprehensive data set on foreign travellers to investigate their happiness at Italian destinations between 2005 and 2014. Using a theory-dependent approach to model happiness at the destination with respect to endogenous and exogenous amenities, personal characteristics and trip features, a great diversity in the mix of amenities affecting tourist happiness is shown. However, some clear spatial patterns emerge. The findings call for place-based policies targeted at the specific needs of each area.

KEYWORDS:

• competitiveness
• endogenous amenities
• happiness
• international tourists
• territorial heterogeneity

JEL:

• C21
• I31
• R11

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1. In line with much of the happiness literature (Bernini & Tampieri, 2019; Kalmijn & Veenhoven, 2005; Veenhoven, 2012), we use the terms ‘subjective well-being’, ‘happiness’ and ‘life satisfaction’ interchangeably. As well underlined by Veenhoven (2012, p. 1), SWB ‘it is an umbrella term for all that is good. In this meaning, it is often used interchangeably with terms such as “well-being” or “quality of life” and denotes both individual and social welfare’.

2. Although a natural way to detect the best-performing local amenities in terms of well-being could be to collect residents’ evaluations from a community, the use of residents’ perceptions may generate biased evaluations (Banzhaf & Walsh, 2008). Choosing residents as judges of the local quality of amenities may lead to a biased judgement, not only because they have chosen their own location based on their idiosyncratic preferences, but also because the results depend on the relationship between the distribution of preferences across several amenities among the residents and the distribution of amenities in the considered territory.

3. Although quality of life can be measured using objective measures, we opted for the subjective approach because of its appealing features. First, SWB measures allow policy-makers to ‘assist individuals in their everyday life decisions, such as where and how to live’ (Diener & Suh, 1997, p. 191) based on personal experience. Second, SWB measures are more flexible than objective measures. Third, SWB measures rely on individual satisfaction as measured by validated items and scales, guaranteeing comparability across communities and over time. Lastly, it has been shown that a model based on perceptions could outperform the models based on measures of objective attributes (Chasco & Le Gallo, 2013).

4. The data used in the maps are provided by the International Tourism in Italy Survey from the Bank of Italy. The survey is presented in the section dedicated to data and descriptive statistics, while data cleaning is discussed in Appendix A in the supplemental data online.

5. In 2011, in Italy, there were 8092 municipalities; however, most were very small (5704 had fewer than 5000 residents) and lacked tourists’ attractions.

6. We made a few changes to this typological classification. In particular, we modified the classification of 19 tourism areas that were classified as ‘other urban destinations’ because 80% or more of the foreign travellers said that they visited those destinations for a specific purpose (10 were changed to seaside destinations, one to mountain destination and eight to cities of art).

7. In the original formulation, each domain was explained by a set of functions DS_j = DS_j(x_j) (j = 1, 2, … , J), where x_j is the sub-selection of x variables for the domain j. In our analysis, we adopted a simplified specification, having information only on the main domains.

8. Using the user-written STATA command ‘gologit2’, we fit a partial generalized ordered logistic model, where the parallel lines constraint was relaxed only for those variables where it was not justified.

9. We computed the marginal effects using the user-written STATA command ‘margeff’, which modifies the calculation of partial effects when sets of dummy variables are included in the model.