This study aims to examine the causal relationship between tourism and overall well-being. The main objective of this research is to inform the policymakers that tourism can play a vital role in shaping the overall well-being in the developing economies.
This investigation used several time series techniques and procedures that include bounds test and autoregressive distributed lag mechanism to analyze the relationship between tourism and overall well-being in Pakistan by using time series data for the period 1980-2016.
The findings suggest a significant positive relationship between tourism and overall well-being both in the short and long run. The authors find that tourism and overall well-being affect each other positively.
This research indicates that policymakers and government can improve the overall well-being through tourism development. However, tourism policies and long-term planning should be focused on sustainable developments for achieving long-term goals. Besides, special incentives should be provided to the private sector for tourism development.
To the best of the authors’ knowledge, this is the first investigation that examines the causal relationships between tourism and overall well-being through objective indicators in a developing economy. This study fills the immense literature gap and provides new directions to scholars to investigate the mentioned relationship through objective indicators.
1. Introduction
Well-being has been a philosophical and sociological concern since the beginning of time, and research has been extended over time to various disciplines such as psychology, health sciences, tourism and economics (Smith and Diekmann, 2017). Tourism studies have become more focused on well-being in the past few decades, both from a theoretical and methodological perspective (Smith and Diekmann, 2017). Extensive research reveals that tourism experiences and activities have a consequence on tourists’ life satisfaction and the well-being of residents. Tourism activities contribute to several life domains such as leisure, self-esteem, self-identity, family life, health and culture (Uysal et al., 2016). Historically, most of the research focused on the impacts of tourism on the resident’s community, such as cultural, environmental, economic, emotional and social (Andereck et al., 2005; Moscardo et al., 2013; Sharpley, 2014; Uysal et al., 2012c; Ap, 1992; Nunkoo and Gursoy, 2012). However, tourism developments also influence quality of life (QOL) or overall well-being of the local community (Liburd et al., 2012; Uysal et al., 2012b; Uysal et al., 2012a) and became a primary concern for community leaders and governments (Aref, 2011; Lipovčan et al., 2014; Aman et al., 2013; Benckendorff et al., 2009; Moscardo et al., 2013). However, these studies have been conducted in developed countries, and less is known about the effects of tourism on well-being in the developing world. This investigation focuses explicitly on tourism’s contributions to the overall well-being in the developing economy of Pakistan.
According to WTTC (2017), tourism’s direct contribution to Pakistan’s economy was $7.6bn (2.7% of total GDP) in 2016 and it suggests an increase of 5.6% per annum from 2017 to 2027. In 2016, the tourism sector in Pakistan directly produced 1,337,500 jobs, 2.3% of total employment, which is expected to rise by 2.5% per annum to 1,757,000 jobs in 2027 (WTTC, 2017). The investment in the tourism sector in 2016 was recorded 9.3% of the total investment, approximately $3.6bn (WTTC, 2017).
Tourism developments improve the quality of residents’ lives by addressing the economic, social, cultural and recreational concerns, and provide certain other benefits (Benckendorff et al., 2009; McCool and Martin, 1994; Peters and Schuckert, 2014). Scholars have used subjective indicators to measure the effects of tourism over well-being (Andereck and Nyaupane, 2011; Aref, 2011; Gjerald, 2005; Khizindar, 2012; Matarrita-Cascante, 2010; Nichols et al., 2002; Kim et al., 2013; Uysal et al., 2012b); the current authors have used objective indicators to measure the effects of tourism on overall well-being in a developing economy. The current literature is scant about the relationship between tourism and overall well-being in developing economies. Since 2005 onwards, tourism and overall well-being rose simultaneously in Pakistan; however, terrorism razes tourism, which is evident from the shocks in tourism activities in Figure 1. The present authors seek to propose a novel perspective of tourism and overall well-being by investigating the causal relationships through economic modeling. The assessment of the current research about tourism and well-being is vital for several reasons. First, this research focuses on the effectiveness of tourism in the enhancement of overall well-being. Second, it assesses the role of overall well-being in supporting tourism activities in a country.
Moreover, this study provides excellent insights to policymakers, politicians and government officials that tourism and well-being contribute to each other. Therefore, policymakers can use tourism development as a policy agenda for the improvements in the overall well-being of general masses. Besides, this investigation contributes to the current literature on the relationship between tourism and well-being in several novel ways. First, to the best of our knowledge, this is the first research that investigates the relationship between tourism and overall well-being in the context of a developing country. Second, we have used objective indicators for examining the relationship between tourism and well-being, while the current literature is mostly focused on subjective measures. Third, we deal with well-being as macro phenomena and our findings suggest that tourism can play a vital role in the enhancement of overall well-being in the developing economies. Fourth, scant studies used econometrics model to map the relationship between tourism and well-being; hence, this study focused on econometrics.
Furthermore, to conduct this investigation, the time-series data of Pakistan from 1980 to 2016 has been used. Several time-series techniques have been applied; for instance, the bounds test approach and the autoregressive distributed lag (ARDL) model. Augmented Dicky–Fuller (ADF), Philips and Perron (PP) and Zivot–Andrew root tests are applied for checking the stationarity of variables and to fulfill the primary assumption of the bounds test approach that all variables should be stationary at I(0), I(1) or a combination of both. Afterward, bounds test approach is used for confirming the long-run relationship between the variables; then, ARDL model is applied for the estimation of long-run relationships. Several diagnostics tests, for instance, Breusch–Godfrey correlation LM test, Breusch Pagan Godfrey heteroskedasticity test, Jarque–Bera normality tests, cumulative sum (CUSUM) and CUSUM square tests have been conducted for checking the stability of the estimated models. The main findings of the study suggest that tourism positively contributes to the overall well-being, while an increase in overall well-being increases tourism activities. This research is composed of six sections; Section 1 deals with the introduction. Section 2 deals with the relevant literature. Section 3 covers the methodology and estimation strategy. Section 4 describes the findings of the study and Section 5 is composed of the discussion. Finally, Section 6 deals with conclusion, practical implications, limitations and future research prospects.
2. Literature review
In the entire world, tourism is considered as a mean of economic development and has become a part of the strategic planning almost in every country (Uysal et al., 2016). Tourism has significant effects on the well-being of the stakeholders (Jurowski et al., 1997). Tourism research has concentrated on the well-being through the associated terms such as life satisfaction, QOL, wellness and happiness (Smith and Diekmann, 2017; de Bloom et al., 2010; Dolnicar et al., 2012; Gilbert and Abdullah, 2004; Neal et al., 2004; Uysal et al., 2016; Smith and Puczkó, 2008; Nawijn, 2011a; Nawijn, 2011b). The tourism industry is facing numerous obstacles such as socioeconomic uncertainty, terrorism, deficient technology, ecological and sustainability issues, and the furthermost risky is the inadequate dissemination of tourism reimbursements.
There is no definite consensus concerning the definition of tourism. But when it comes to explanation within the basic terms, it can be summed up that:
Tourism is a collection of activities, services, and industries which deliver a travel experience comprising transportation, accommodation, eating and drinking establishments, retail shops, entertainment businesses and other hospitality services provided for individuals or groups traveling away from home (Robert and Goeldner, 1986).
The mentioned components in the above definition directly or indirectly contribute to the well-being of the tourists and destination residents.
Prominent scholars such as Uysal et al. (2012a), Uysal et al. (2012b) and Uysal et al. (2012c) proposed two models for examining the relationship between tourism and well-being from a system point of view; the first model is based on variables that impact the well-being of tourists and the second focuses on the factors that influence the well-being of the host communities. The key focus in the first model shows tourists’ experiences that contribute to QOL (Dann, 2012), satisfaction with life domains and overall well-being (Sirgy et al., 2011). The second model illustrates how tourism influences the living conditions of the host community at a destination by affecting economic conditions, infrastructure, life amenities and destination competitiveness; in fact, it contributes to the overall life domains of the community as a whole (Crouch and Ritchie, 2012).
Research suggests that tourism contributes to residents and tourists well-being, for instance, studies suggest that tourism activities refresh mind and body (Bushell and Sheldon, 2009), improve awareness of natural and cultural heritage, maximize self-esteem (Hartwell et al., 2018) and help in personality development and self-identity (Hartwell et al., 2018; Dolnicar et al., 2012). The research on tourism contribution to the resident community is examined through the impacts on QOL and its consequences toward tourism development (Jamal and Dredge, 2014; Sharpley and Telfer, 2014; Andereck et al., 2005). The trends suggest that the impacts of tourism vary according to the level of individual contact, tourism development stage and benefit obtained from tourism (Hartwell et al., 2018). Wiseman and Brasher (2008) suggest that community well-being and its dynamics can be better understood by community engagement and policy paradigms at a destination. Besides, scholars pointed out that traditional approaches of measuring well-being has been proved methodologically problematic (Hartwell et al., 2018) and offered new conceptual paradigms for measuring specific dimensions and values that can contribute to the life domain of the destination community (Tyrrell et al., 2013; Tyrrell et al., 2010). What is debatable from these studies is that the prevailing methodologies of measuring tourism contribution to the community well-being is questionable and required a mix of subjective and objective measures that can reconcile the deficiencies of each other. Unfortunately, objective side of measuring tourism impacts on community at large has been ignored. Hence, this research focuses on the objective indicators to measure the tourism influence on community at macro level.
It is evident from various studies that governments, politicians and community leaders consider tourism as a substantial instrument for enhancing employment prospects, revenues, social interrelation and economic diversity (Kim et al., 2013). The trends suggest that tourism has both progressive and adverse impacts on well-being (Kim et al., 2013; Allen et al., 1988; Prentice, 1993; Tosun, 2002; Deaton, 2008). Researchers have examined the effects of tourism on social well-being (McCabe et al., 2010; Coulthard et al., 2011; Morgan et al., 2015). Additionally, scholars have captured the influence of tourism on cultural well-being (Uysal et al., 2016; Kousis, 1989; Hall and Brown, 2006; Kim et al., 2013) and on environmental well-being (Butler, 1999; Holden, 2003; Andereck, 1995; Farrell and Runyan, 1991). Scholars have also investigated the rapport of tourism and material well-being (Konu, 2010; León, 2007; Pesonen and Komppula, 2010). The study by Kim et al. (2013) expresses that tourism influences residents’ material as well as non-material well-being. The research by Woo et al. (2015) conveys that tourism development has impacted both material and non-material well-being of the residents. Trends suggest that a mix of objective and subjective indicators such as poverty, per capita income, crime rate, pollution and perceptions have been used for measuring well-being (Kim et al., 2013; Crotts and Holland, 1993). Tourism activities contribute to numerous life spheres such as household, social, leisure, economic and cultural aspects of life (Beeton, 2006).
A number of countries are developing their rural areas by means of tourism (Ashley et al., 2000; Roe et al., 2004; Beeton, 2006) and empowering the less-privileged class of the society by providing occupational openings that support the overall well-being (Chant, 2005; Cukier et al., 1996; Roe et al., 2004) and enrich self-esteem (Gu and Ryan, 2008; Wang and Xu, 2015). Tourism improves the standard of living and offers avenues for employment (Ko and Stewart, 2002; Upchurch and Teivane, 2000). Scholars submit that tourism growth enhances cultural exchanges and provides leisure prospects for natives and tourists (Stylidis et al., 2014; Dyer et al., 2007).
Studies reveal that tourism development contributes to the economic prosperity and well-being of the community (Raymond and Brown, 2007). Moreover, in distant geographies of several countries, tourism has enhanced the socio-economic well-being of the residents (Briedenhann and Wickens, 2004; Oakes, 2005). In most of the developing countries where tourism development is in the nascent stage, it has been regarded as a positive contributor to the overall well-being (Kim et al., 2013; Uysal et al., 2012a; Sharpley, 2014; Woo et al., 2015; Uysal et al., 2012c). However, some of the related works are summarized in Table 1.
Summary of related literature
| Authors | Measurement | Summary |
|---|---|---|
| Perdue and Gustke (1991) | Economic growth, education, population growth and health facilities were used to measure the influence of tourism residents’ QOL | The summary of the study suggests that tourism development has increased the material, social and health conditions in the community |
| Crotts and Holland (1993) | Objective indicators were used to measure tourism impact on well-being | This study suggests that tourism increased residents’ income and housing costs at the destination |
| Jurowski et al. (1997) | Subjective indicators were used for developing a theoretical framework by applying path analysis | This study applied the social exchange theory to measure tourism impacts and developed a theoretical framework. This study suggested that several factors, i.e. level of economic gain, use of resources, attitude and community attachment shapes residents’ perceptions and their support for tourism development |
| Bachleitner and Zins (1999) | Subjective indicators were used to analyze how cultural tourism influence residents QOL | Using the TIAS model developed by Lankford and Howard, an empirical analysis was conducted. The four-quadrant model of social impacts by Bjorklund and Philbrick was applied to explain perceptual shifts. The findings suggest that cultural tourism enhanced tourism growth and in rural communities. It is noted that cultural tourism has great influences on the socio-psycho behaviors of individuals in the community |
| Andereck and Vogt (2000) | Subjective indicators were applied to measure tourism development and support for tourism development with 41 items with a five-point Likert scale | The main objective of this article is to investigate the relationship between tourism and residents’ support for tourism development. Seven communities were identified for the research. The findings suggest that communities differ concerning support for tourism development and positive perception of tourism |
| Spiegel et al. (2007) | This study used focus groups and critical interviews in two coastal communities | Study to determine health impacts on residents of the expansion of tourism in Cuba |
| Wheeler and Laing (2008) | A mix of qualitative and quantitative data was used to measure the level of satisfaction | The findings suggest that the relationship between tourism and well-being was moderated by livability |
| Vargas-Sánchez et al. (2009) | Six objective indicators were used to measure community overall well-being | The findings of this research suggest that overall community satisfaction was positively influenced by positive impacts of tourism. However, individual benefits and negative perception does not influence community satisfaction |
| Meng et al. (2010) | A total of 17 objective indicators were used to measure the QOL | The findings suggest that people living in tourism-developed areas were living better than those who were living in less-developed tourism destinations |
| Aref (2011) | A questionnaire survey was conducted on a Likert scale from 1 to 5 | The findings suggest that tourism positively affects resident well-being in Sheraz, Iran |
| Nawijn and Mitas (2012) | A self-administered questionnaire was used to measure residents’ attitude toward tourism and its impacts of community well-being | The findings suggest that perceived tourism impacts are related to life satisfaction. Moreover, tourism positively influences health, infrastructure, personal relationship and services |
| Khizindar (2012) | Three subjective indicators were used to measure the QOL | The findings suggest that social, environmental and cultural impacts of tourism influence resident’s well-being |
| Kim et al. (2013) | This study used nine subjective indicators to measure several aspects of well-being and overall life satisfaction | The results of this article suggest that stages of the tourism life cycle moderate the relationships between tourism and well-being |
| Angeloni (2013) | Secondary data was used to analyze the weaknesses of cultural tourism in Italy | The findings suggest that cultural tourism can enhance the well-being of locals by enhancing the function of local government. The author argues that a destination that manages its resources efficiently and effectively has the ability to improve the QOL of the community |
| Buzinde et al. (2014) | This article applied a bottom-up approach to investigate original elements of well-being for a better understanding of how tourism affects indigenous experiences of well-being | Two focus group discussions were undertaken in the two communities to record the basic understanding of well-being in the community. Besides, what factor influence well-being and how tourism influences the environment, culture and economy at Maasai and what are the challenges and positive outcomes of tourism. All ages of people agreed that tourism benefits the local community, but it has negative impact on community well-being |
| Lipovčan et al. (2014) | Subjective well-being measured on a scale of 0-10 was used to measure tourists’ satisfaction and residents’ well-being at 41 different destinations. The destinations were grouped into three categories based on touristic quality | The residents of destinations with the higher quality of tourist products were happier and satisfied with their lives than those living in the destinations with medium and lower quality of tourist offers |
| Sharpley and Telfer (2014) | This review article explores the development of research into residents’ perceptions of tourism | The research presents a critical review of relevant research related to tourism development and well-being. The article provides a good overview of the progress made in tourism research and also criticizes and provides suggestions where required |
| Woo et al. (2015) | Six subjective indicators were used to measure tourism development and its impacts on QOL | Tourism development positively affects material and non-material life that contribute to QOL. However, QOL is a determinant of tourism future developments |
| Jeon et al. (2016) | Four subjective indicators were used to measure resident’s QOL | The article suggests that resident QOL was positively influenced by perceived economic benefits, social costs and environmental sustainability |
3. Methodology
This study explores the relationships between tourism and overall well-being in a developing economy. Annual time-series data of Pakistan has been collected from the World Bank online depository and different versions of Pakistan statistical yearbooks for the period 1980-2016. In this research, the literacy rate represents social well-being (LNSWB); life expectancy represents health well-being (LNHWB), Human Development Index (HDI) represents overall well-being (LNOWB), and tourism receipts (LNTOUR) represent tourism activities. A dummy variable is used for measuring terrorism shocks. The data has been transformed by taking the natural log of all variables. The details of all variables are provided in Table 2.
Variables details
| Variables | Description | Proxy | Units |
|---|---|---|---|
| LNOWB | Overall well-being | HDI | Score (0-1) |
| LNHWB | Health well-being | Life expectancy | In years |
| LNSWB | Social well-being | Literacy rate | Per population |
| LNTOUR | Tourism activities | Tourism receipts | In million $ |
Scholars and international organizations such as Katz et al. (1983), Council (2011) and Robine and Ritchie (1991) have used life expectancy as an overall measure of population health. The literacy rate has been regarded as a direct measure of life achievements and access to the minimum level of education; HDI is a compound indicator of overall well-being (Cooke et al., 2007). International organizations such as Organisation for Economic Co-operation and Development also used the same type of indicators to measure well-being (Beaumont, 2011).
Scholars have applied ARDL model on panel and time-series data in tourism research (Lin et al., 2015; Liu and Pratt, 2017; Katircioglu, 2009; Adnan et al., 2013). This research used the ARDL model to estimate the long- and short-run relationship between tourism and overall well-being. Health and material well-being were taken as control variables. Besides, ARDL and other statistical tools are used to analyze well-being, tourism, economic growth and other factors (Gebrehiwot, 2016; Ahmad and Riaz, 2011; Shahbaz and Aamir, 2008; Lee et al., 2013; Ridderstaat et al., 2016). The following equation (1) and (2) are assumed to analyze the relationship between the mentioned variables.
Where β0 is constant; β1, β2, β3 and β4 are the parameters; and µt is error term, assumed to be normally distributed.
3.1 Estimation strategy
This investigation used the ARDL bounds testing approach to co-integration developed by Pesaran and Shin (1998) and Pesaran et al. (2000). Traditional approaches of co-integration such as Johansen and Juselius (1990), Phillips and Hansen (1990) and Engle and Granger (1987) have certain demerits in comparison to ARDL bounds test approach. The traditional approaches can be used only when variables are stationary at the order I(1) while the ARDL bounds test approach can be used whether the variables are stationary at I(0) or I(1), or a combination of both. The ARDL approach is appropriate for the analysis of the present data because of small sample size (Nkoro and Uko, 2016). The ARDL error correction model accommodates satisfactory lags that capture the data generation process in general to specific framework (Laurenceson and Chai, 2003). Under the ARDL approach, equations (1) and (2) are molded as under for the ARDL:
In equations (3) and (4), the parameters β1, β2, β3, β4 and β5 represent short-run measurements and β6, β7, β8, β9 and β10 are the long-run elasticities. Moreover, Δ is the first difference operator and q indicates the optimal lag length. Error correction terms (ECTs) of equations (3) and (4) are given below:
In equations (5) and (6), q represents the optimal lag length, γ is the speed of adjustment parameter and ECT represents the error correction term, derived from long-run relationships. The subsequent step in the ARDL approach is to calculate F-statistic and to equate it with tabulated critical bounds to examine whether the long-run co-integration exists or not (Pesaran et al., 2001; Shahbaz, 2013). Besides, a dummy variable is used for measuring the effects of terrorism from 2002 to 2013.
The null hypotheses for equations (3) and (4) can be stated as H0: β6 = β7 = β8 = β9 = β10 = 0, which assume that no long-run relationship between the variables exists, whereas the alternative hypothesis for co-integration can be Ha: β6 ≠ β7 ≠ β8 ≠ β9 ≠ β10 ≠ 0. If the F-statistic calculated value is larger than the upper bounds (F > FU) at 1%, 5% or 10% significance level, the decision will be in favor of co-integration, if it falls shorter than the lower bounds (FL > F), there will be no co-integration and in case F-statistic falls between lower and upper bounds (FL < F< FU), the result will be called inconclusive. ARDL model assumes that variables should be stationary of I(0) or I(1) or the combination of both; three different unit root tests are applied to authenticate whether the variables are stationary or not, i.e. ADF test (Dickey and Fuller, 1979), PP test (Phillips and Perron, 1988) and Zivot–Andrews test (Zivot and Andrews, 1992). ADF test is based on the following equation:
Whereas the PP unit root test can be illustrated algebraically as:
4. Findings
The Jarque–Bera statistics in Table 3 confirm that all the variables are normally distributed. The normality specifies the authenticity of data and allows us for further analysis.
Descriptive statistic
| LNOWB | LNSWB | LNHWB | LNTOUR | |
|---|---|---|---|---|
| Mean | −0.808 | 1.462 | 4.129 | 5.146 |
| Median | −0.823 | 1.435 | 4.132 | 5.123 |
| Maximum | −0.597 | 1.887 | 4.196 | 5.910 |
| Minimum | −1.022 | 1.011 | 4.048 | 4.335 |
| Std dev. | 0.134 | 0.297 | 0.044 | 0.438 |
| Skewness | 0.084 | −0.017 | −0.179 | 0.114 |
| Kurtosis | 1.737 | 1.667 | 1.888 | 1.977 |
| Jarque–Bera | 2.432 | 2.666 | 2.044 | 1.647 |
| Probability | 0.626 | 0.263 | 0.359 | 0.438 |
The outcomes of ADF and PP tests in Table 4 reveal that the variables are stationary either at I(0) or at I(1). The output recommends that the ARDL bounds test approach is appropriate for estimating the relationship between the mentioned variables. Scholars have criticized ADF and PP tests that they do not provide any information about structural breaks, which may cause biasedness (Baum, 2003).
Zivot–Andrews unit root test provides evidence about structural breaks; hence, Zivot–Andrews unit root test is used for checking the robustness of the results in Table 3. The findings of the Zivot–Andrews test are given in Table 5.
Zivot and Andrews’ (1992) unit root tests
| Unit root at levels | Unit root at first differences | |||
|---|---|---|---|---|
| Variable | t-statistic | Year of break | t-statistic | Year of break |
| LNOWB | −5.330* (0, I) | 2005 | −3.122**(1, I) | 2008 |
| LNSWB | −3.459** (0, I) | 2004 | −6.911* (0, I) | 1990 |
| LNHWB | −1.633 (0, I) | 1987 | −6.449**(0, I) | 2008 |
| LNTOUR | −3.588** (0, I) | 2004 | −7.831* (0,I) | 2000 |
Notes:
*, ** and *** refer to rejection of null hypothesis that unit root exist at 1%, 5% and 10%, respectively. () refers to optimal breaks
Feridun and Shahbaz (2010) suggest that F-statistic is sensitive to the lag order of the variables. Hence, appropriate lag length selection is essential for applying the ARDL approach. The smaller the value of the criteria, the better will be the model. In the case of current data, the Akaike information criterion shows the minimum value at lag 4; hence, the optimal lag length is 4. Bounds test results are summarized in Table 6; the computed F-statistic values for Models 1 and 2 are greater than the upper bound at 1% significance level. This confirms the co-integration between the variables, which authenticate long-run relationships between overall well-being and tourism for the period 1980-2016.
The long-run relationships for both models are summarized in Table 7. The results of ARDL Model 1 with optimal lags (1, 4, 4, 2, 2) recommend that a positive relationship exists between tourism and overall well-being. The estimates of ARDL Model 1 suggest that 10% increase in tourism increases overall well-being by 0.42%.
The estimated long-run coefficient results (ADRL) estimation
| Model 1 (1, 4, 4, 2, 2) | ||||||
|---|---|---|---|---|---|---|
| Determinants | Constant | LNTOUR | LNSWB | LNHWB | ||
| Coefficient | −8.882 | 0.042 | 0.112 | 1.875 | ||
| t-stat | −4.501* | 5.940* | 1.411 | 3.756* | ||
| Model 2 (2, 3, 4, 4, 2) | ||||||
| Determinants | Constant | LNOWB | LNSWB | LNHWB | ||
| Coefficient | 266.63 | 20.485 | 0.796 | 59.733 | ||
| t-stat | 5.814* | 4.318* | 0.348 | 5.483* | ||
| Diagnostic tests | ||||||
| Parameters | Model 1 | Model 2 | ||||
| Adjusted R2 | 0.822 | 0.674 | ||||
| 1.73 (0.420) | 0.024 (0.988) | |||||
| 12.87 (0.744) | 18.25 (0.506) | |||||
| 5.55 (0.07) | 5.19 (0.076) | |||||
| F-statistic (Prob) | 9.435 (0.001) | 4.37 (0.006) | ||||
| D/W statistics | 2.24 | 2.62 | ||||
Notes:
*, ** and *** refer to level of significance at 1%, 5% and 10%, respectively. () represent probability
However, the findings of Model 2 with optimal lags (2, 3, 4, 4, 2) suggest that overall and health well-being are positively associated with tourism. The results of Model 2 recommend that a 1% increase in overall and health well-being increases tourism by 20.48% and 59.73%, respectively. All the diagnostic tests confirm that no violation of any linear regression assumptions occurs. This indicates that models are free of specification problems.
The short-run elasticities and ECTs are reported in Table 8. The ECT specifies the speed of adjustment toward the long-run equilibrium. The ECT coefficient demonstrates how speedily variables return to equilibrium; it must be significant at 5% level, having a negative sign (Pahlavani et al., 2005). Banerjee et al. (1998) suggest that a high value of ECT is further proof of a long-run stable relationship. ECT values for Models 1 and 2 are −0.777 and −0.886, correspondingly significant at 1%, proposing that a deviation from equilibrium will be corrected in the future by 77.7% and 88.6%, respectively. The calculated short-run elasticity for tourism has a significant positive relationship with overall well-being in Model 1; the coefficient of tourism indicates that a 10% increase in tourism will enhance overall well-being by 0.21% within one year. Moreover, the dummy variable is negative in the short run, indicating that terrorists’ activities have negative effects on the overall well-being. Furthermore, the short-run dynamics of Model 2 suggest that overall and health well-being are positively associated with tourism. The short-run coefficients of Model 2 suggest that 1% of overall and health well-being increase tourism by 17.12% and 69.86%, respectively.
Error correction model representation of the selected ARDL models
| Variables | Model 1 Δ LNOWB | Model 2 Δ LNTOUR |
|---|---|---|
| Δ LNOWB | 17.126 (3.327)* | |
| Δ LNSWB | 0.069 (1.847)*** | 1.689 (1.651) |
| Δ LNHWB | 2.615 (2.272)** | 69.865 (1.831)*** |
| Δ LNTOUR | 0.021 (4.008)* | |
| Δ Dummy | −0.008 (−2.597)** | −0.221 (−1.622) |
| ECT t(−1) | −0.777 (−5.738)* | −0.886 (−4.923)* |
Notes:
*, ** and *** refer to level of significance at 1%, 5% and 10%, respectively. () represent t-statistic
The graphs of CUSUM and CUSUM square presented in Figures 2-5 reveal that plots for both CUSUM and CUSUM square are between the critical boundaries at a 5% level of significance. The results of CUSUM and CUSUM square authenticate the accuracy and stability of long- and short-run parameters of the ARDL models.
5. Discussion
The World Tourism Organization proposes tourism as a tool through which the Sustainable Development Goals (SDGs) can be accomplished (UNWTO and UNDP, 2017). Well-planned tourism helps to improve human well-being by reducing poverty, hunger, ensure education for all and sustainable development. Literature trends reveal that Tourism has a great potential to accelerate progress across the SDGs (UNWTO/GTERC, 2017; Adnan et al., 2013). The tourism sector can generate quality jobs for durable growth, reduce poverty and offer incentives for environmental conservation (UNWTO and UNDP, 2017). The findings of this study indicate that tourism has significant positive effects on overall well-being both in the short and long run. A 10% increase in tourism activities increases overall well-being by 0.42% and 0.48%, respectively, in long and short run. Hence, a focus on sustainable tourism development in the developing economies would be beneficial for economic growth, poverty alleviation and improvement in education. Our proxy HDI for overall well-being is a combined indicator of economic growth, life expectancy and access to education; therefore, tourism development in the developing nations would collectively provide support directly or indirectly to the pillars of well-being. Through the development of tourism, Pakistan will be able to achieve some of the SDGs, either directly or indirectly. The results also suggest that an increase in the overall well-being increases tourism activities; 1% increase in overall well-being increases tourism by 20.48% and 17.13%, respectively, in the long and short run. Several scholars argued that poverty, hunger and deficiencies in education facilities significantly contribute to crimes, social evils and terrorists activities in a country (Alcalá et al., 2017), which are the major obstacles for tourism development; hence, rise in overall well-being would help Pakistan to overcome the social evils prevailing in the country. Thus, tourism development can improve the well-being of the people, and such improvement will accelerate tourism activities as suggested by our results. Moreover, the findings are in line with the results of several scholars who suggested that tourism enhances local communities’ well-being, such as Crouch and Ritchie (1999) and Slee et al. (1997).
6. Conclusion
Debate on the relationship between tourism and well-being has produced a vast literature, but the results are still inconclusive and remained open for discussion. To supplement the previous findings, we attempted to capture the relationship between tourism and overall well-being by using econometric modeling for a developing country. This paper sets out to explore the empirical relationship between tourism and overall well-being. In doing so, it was revealed that tourism and overall well-being have significant impacts on each other. The existing literature suggests that tourism contributes to the economic, social, cultural, environmental and emotional aspects of the host community (Adnan et al., 2013; Allen et al., 1988; Uysal et al., 2012a). Pakistan is a country of natural, cultural and heritage wonders. A well-thought tourism development would be helpful in the enrichment of job creation, tourism SMEs development and economic growth; moreover, it would be supportive in the achievement of SDGs. The quantitative estimates of our models suggest that tourism development in Pakistan would enhance the living standards. Though previous studies through subjective indicators have suggested that tourism can improve well-being, but to the best of our knowledge, scant literature is available on objective indicators. Besides, this research suggests that well-being positively contributes to tourism; hence, it provides evidence for a bi-directional causal relationship between tourism and overall well-being. This study contributes to the current literature on the relationship between tourism and overall well-being in several ways. First, this study investigates the relationship between tourism and well-being on a macro level. Second, we used objective indicators instead of subjective, hence, provide a new methodological approach. Third, we found that tourism and well-being are bi-directional. Fourth, tourism development can be used as a strategic tool to uplift the well-being in developing countries. Therefore, in light of the bi-directional relationship between tourism and overall well-being, it is suggested that both sides of the equation positively contribute to each other, thus, better tourism and welfare policies at national level will improve the living standards of the residents and also increase tourism activities.
6.1 Practical implications, limitations and future prospects
This investigation is of particular importance for policymakers, community leaders and politicians in Pakistan. Around the world, the private sector plays a vital role in tourism development; hence, Pakistan should provide finances to private sector for the required development. The institutional mechanism should be revised for sustainable tourism development to achieve maximum SDGs. Tourism policymakers should strengthen and facilitate the dialogues between the related stakeholders for taking advantage of tourism’s inter-linkages to have better impact on well-being. A better financial system should be developed for the development of tourism SMEs in rural areas of Pakistan. Academia should be involved and funded for in-depth tourism research for the development of knowledge, strategies and research-based policies. There are certain limitations to the current study. First, the outcomes may correctly address the case of some developing countries but may not be fully applicable to other nations. Second, it provides a macro view of the subject mentioned, which may or may not be true at a micro-level. In future, scholars can find the relationship between tourism and overall well-being by using panel data. This research puts a call for scholars to further research the relationship between tourism and well-being by using different combination of objective indicators.
