This paper estimates a total factor productivity index that allows for a rich decomposition of productivity in the tourism industry. We account for two important characteristics: First, the heterogeneity between multiple tourism destinations, and second, the potential endogeneity in inputs. Importantly we develop our index at the macro level, focusing on cross-country comparisons. Using the Bayesian approach, we test the performance of the model across various priors. We rank tourism destinations based on their tourism productivity and discuss the main sources of productivity growth. We also provide long-run productivity measures and discuss the importance of distinguishing between short-run and long-run productivity measures for future performance improvement strategies. © ۲۰۱۷ Elsevier Ltd. All rights reserved.

Introduction

“Productivity isn’t everything, but in the long run it is almost everything. A country’s ability to improve its standard of living over time depends almost entirely on its ability to raise its output per worker” (Krugman, 1994, p.9). Despite being a high priority on the World Tourism Organization (UNWTO) research agenda, the productivity analysis of the tourism industry has not received much attention in the tourism literature. There is a continuous effort at most tourism destinations to strengthen the productivity of their tourism industry (Cvelbar, Dwyer, Koman, & Mihali
c, 2016). As stated by Assaf and Dwyer (2013, p.1234), with the tourism industry often perceived as a low productivity industry, productivity analysis is “crucial to evaluating tourism sustainability and reshaping tourism activities. There is a direct link between productivity and profitability, as when productivity increases, the tourism industry’s competitiveness in labour, capital and real estate markets also increase”. The tourism competitiveness literature also highlights the important link between competitiveness and productivity. Dwyer, Forsyth, & Rao (2000, p. 9), for instance, view competitiveness as “a general concept that encompasses price deferential coupled with exchange rate movements, productivity levels of various components of the tourist industry and qualitative factors affecting the attractiveness or otherwise of a destination”. Echoing this, Crouch and Ritchie (1999, p.149) have emphasized that ensuring higher destination productivity an effectiveness necessitates from each destination management organization (DMO) “the responsibility to disseminate key market and performance information to its members on a timely basis”. Even competitiveness at the firm level can be enhanced through productivity improvements (Dwyer & Kim, 2003). While some research evaluated competiveness from the perspective of productivity, the two are often viewed as separate but related components (Assaf & Josiassen, 2012). The concept of “competitiveness” should not also be used to reflect the productivity of the tourism industry (Assaf & Josiassen, 2012) – productivity is a major driver of “competitiveness”, and not “competitiveness” itself (Cvelbar et al., 2016). Often misleading is the definition of productivity in the tourism industry. The various league tables providing productivity indicators of the tourism industry “neither takes explicit account of productivity in tourism” (Blake, Sinclair, & Soria, 2006, p. 1100). Productivity is a complex phenomenon and involves several components; hence using simple metrics to reflect the overall tourism productivity can be misleading for policy implications (Barros, Botti, Peypoch, Robinot, & Solonandrasana, 2011). Over the last decade, there has been an increasing focus on analysing the performance of the tourism industry using the concept of “technical efficiency” (Assaf & Josiassen, 2012; Barros et al., 2011; Peypoch & Solonandrasana, 2006). However, while technical efficiency is a comprehensive measure of performance, it is only one component of productivity-productivity growth is not driven by technical efficiency alone, but by other factors such as “innovation” and “output growth” (Coelli, Rao, O’Donnell, & Battese, 2005).