The asymmetric impact of air transport on economic growth in Spain: fresh evidence from the tourism-led growth hypothesis

ABSTRACT

The tourism sector has emerged as an essential driver for economic growth strategies during the last decades. An asymmetric long-run effect of air transport on economic growth is validated assuming a process of social globalization in Spain between 1970 and 2015. To achieve the study’s objective, the recent asymmetric autoregressive distributed lag methodology framework advanced by Shin, Yu, and Greenwood-Nimmo (2014) is applied. For determining the causality direction, this methodology is applied in conjunction with the non-parametric causality test proposed by Diks and Panchenko (2006). The current study also accounts for the effects of renewable energy use and urbanization process over economic growth. Empirical results showed that air transport, urbanization process and social globalization exert positive and significant implications over economic growth, while renewable energy use reduces economic growth, as consequence of an energy mix sustained by fossil sources. Based on these outcomes several policy recommendations were offered in the concluding section.

KEYWORDS:

• Air transport
• renewable energy use
• social globalization
• tourism development
• economic growth
• cointegration analysis

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Air transportation (AT) in this study context refers to the movement of persons, customer’s cargo by planes, aircrafts. AT in recent times has evolved to be main/preferred sources of movement given in unique traits of comfort and speed in the transportation sector. This study uses AT as to proxy tourism after studies of (Brida et al., 2018; Husein and Kara, 2020).

2 For brevity, details on other related literature are presented in in appendix section.

3 For brevity, the results of the BDS test can be made available upon request.

4 Even some academics recommend that there is no need for stationary checking for the ARDL method (Ibrahim, 2015), ARDL contains one limitation; when any series in model is stationary at second difference I(2), ARDL cannot be employed, becoming F-statistics value invalid (see Ibrahim, 2015). Therefore, to leave second difference it is recommended to use Dickey–Fuller (ADF) test. Our study applies ADF with structural brake, reported in Table 1.

5 See, Zivot and Andrews (1992), Banerjee, Lumsdaine, and Stock (1992) and Vogelsang and Perron (1998).

6 See, Fox (1972) and Tsay (1988).