Economic impact of Chinese tourism on Australia

Abstract

China’s trade, tourism and limited foreign direct investment (FDI) to Australia have been regarded as playing an important part in Australia’s growth and prosperity in recent years. In spite of the fact that these activities are the three principal growth determinants in modern economic integration theory, growth studies based on this theory’s structural framework, while highly appropriate, have hardly been undertaken. This article proposes to fill the gap by formally developing an endogenous causal model of simultaneous growth and tourism for policy analysis. In this model, trade, FDI and tourism are specified as the main contributing factors to growth. Simultaneously, gravity theory (including growth) and the Ironmonger–Lancaster new consumer demand theory determine tourism, while ‘economic conditionality’ potentially affecting both growth and tourism in the sense of Johansen is recognized and incorporated. The model is then applied to Australian and Chinese data for the important post-Japanese tourist boom period 1992–2015, to provide substantive findings on three questions: the impact of Chinese tourism to Australia, Chinese tourism determination and the effects of Chinese trade and key macroeconomic indicators on Australian economic growth. Significant policy implications are then developed for use by government tourism planners and policymakers.

Keywords

Australian economic growth China’s trade and tourism economic integration theory endogenous growth

Introduction

Tourism has been an important contributing factor to gross domestic product (GDP) growth in world-wide economies. China, with its recent rapid increase in economic growth, increasing living standards and wide-ranging globalization reforms, has provided a large supply of tourists to the world, and this is especially the case for both geographically and culturally attractive tourist destinations such as Australia. In the case of Australia, not only has the number of tourists from China exponentially increased since the 1970s (following the Japanese tourist boom of the 1970s and 1980s), its bilateral trade and more limited foreign direct investment (FDI) (restricted by local government policy) have also greatly expanded (ABS, 2017). The economic benefits from these activities not only have helped to boost the health of the Australian economy, but also boosted its sustained growth and stronger political ties and closer cooperation with its trade partner.

Beginning with the pioneering work of Guthrie (1961), a large literature has developed in the last three decades dealing with the economics of tourism, with diverse focus and methodology (see, e.g. Song et al., 2012, for a comprehensive review).

However, with the intense process of increasing globalization and economic integration (EI) in more recent years (WTO, 2017), international attention by policymakers and scholars has been particularly focused on the field of regional trade agreements (RTAs) where major activities such as trade, FDI and services (including tourism) have been postulated as the key contributing factors to growth and development, thereby improving living standards and alleviating poverty especially in developing economies. In spite of these important developments and growing policy relevance, studies of trade, FDI and tourism in this new structural framework have not been widely undertaken, with few exceptions in trade, official development assistance and climate change fields such as Tran (2004), Tran (2007) and Tran and Limskul (2013).

This article advances research by addressing this gap in the literature in the field of tourism economics, focussing on the simultaneous endogenous tourism and growth relationship, with the impact of China’s tourism on Australia as a study focus. The plan of the article is as follows. The ‘Trend and pattern of Australia’s main growth indicators’ section provides a review of the trends and patterns of the main indicators of Australia’s growth and China’s trade in tourism and FDI to Australia, during the volatile period of 1990–2015. The ‘A new approach to endogenous growth and tourism modelling’ section develops a new model of endogenous growth and tourism based on the EI or RTA theory, gravity theory, Ironmonger–Lancaster consumer demand, and Johansen (1982) factors which are fundamental to determining the effects of China’s tourism to Australia. The ‘Empirical implementation and substantive findings’ section describes the data and presents the empirical findings from the model’s estimation using a systems approach, a three-stage least squares (3SLS), and measurements of statistical reliability. The ‘General findings and policy implications’ section describes policy implications and the ‘Conclusion’ section concludes.

Trend and pattern of Australia’s main growth indicators

The trends of GDP growth (rate of change of real GDP, base = 2010) for Australia and China are compared in Figure 1 for the period 1990–2015. From Figure 1, we note that while starting at almost the same rate of growth in 1990 (at 3.08% for Australia and 3.93% for China), China’s economic achievement (due to its increasing openness and economic and social reforms, accompanied by increasing global economic and political influence) surpassed Australia in the following period, with the biggest gap in 1992 (post-Gulf War) and 2007 (pre-global financial crisis (GFC)). Although China’s growth slowed and Australia’s improved after 2010, the gap was still significant in 2015 at 2.59% for Australia versus 6.93% for China. On average, annual growth was 3.06% for Australia and 9.72% for China. Interestingly, both countries’ economic performance seemed to be affected by the GFC, but China, a major transition economy in Asia, was not impacted by the Gulf War in 1991 and the terrorist attacks in New York in 2001.

Figure 1.

Australia’s and China’s growth (%) and China’s tourism (the annual number of China’s tourists in ‘000), 1990–2016. YC and YCCN denote Australia’s and China’s annual growth, respectively (primary axis), and TCNT denotes Chinese tourism (secondary axis). Source: ABS (2017) and ERS-USDA (2017).

Figure 1 also describes the annual number of China’s tourists (in ‘000) arriving in Australia during 1991–2016. As observed, the data seem to support Australia’s tourism agencies and operators’ great optimism with exponential growth in recent years, leading to optimistic future planning and development. More specifically, from a small number of 16,500 Chinese tourists in 1991 to 1,199,100 in 2016, the trend appears to also steadily rise to 2009 and then grow more rapidly from 2010. Interestingly, this trend was observed even during the period when China’s growth was slowing (see Figure 1). In regard to regional and global crises, the trend for Chinese tourists arriving in Australia was apparently adversely affected by the GFC in 2009, but less so by the Iraq War in 2003.

In contrast with the fast rising trend of Chinese tourists coming to Australia since 1990, its trade flows, especially overall FDI (Chinese FDI to Australia data were limited due to Australia’s restrictive investment review board policy, and not available for the period under study) as a percentage of GDP have more modest growth (refer to Figure 2). In 1990, the trade ratio (or openness) was only 0.68% (reflecting relatively low openness or low trade liberalization) and 1.36% for FDI (reflecting relatively low attraction to foreign investors). In 2015, these figures were 9.02% and 2.28%, respectively, reflecting high rising trade and moderately rising FDI inflows. However, there is an indication that China’s trade flows to Australia have been rising since 1990, and started rising more noticeably since the terrorist attacks in New York in 2001, which is also the year of China’s ascension to the WTO. On average, the annual trade ratio was 4.14% and for FDI it was 1.53% for the whole period. In contrast, Australia’s annual trade ratio with the rest of the world was slowly declining from 2001 and was as high as 26.79% on average. However, China is Australia’s biggest single trade partner in recent years.

Figure 2.

Australia’s trade with China, others, and FDI/GDP, 1990–2015 (%).TCNY, Australian trade with China/GDP; T0CNY, Australian trade with the rest of the world/GDP; FDIY, FDI/GDP (secondary axis). GDP: gross domestic product; FDI: foreign direct investment. Source: ADB (2017).

Two main economic factors commonly stated to contribute to international tourism are the destination’s cost of living (inflation affected) and the origin’s real exchange rate (Gerakis, 1965) or purchasing power. These data (base year 2010) are given in Figure 3 and we note the fairly stable real exchange rate movement in China and the high degree of volatility of inflation in Australia especially in the 1990s. Figure 3 indicates that China’s real exchange rate peaked during the country’s major devaluation period in 1994, which many suggested (e.g. Tran, 2002b) generated contagion effects leading to the Asian financial crisis (AFC) in 1997. The rate has been falling in recent years partly due to international pressure on China to curb its high surplus balance of trade. Australia’s inflation was interestingly affected by the AFC of 1997 and slightly less by the GFC in 2009. However, Australian inflation has been slowly declining since 2001.

Figure 3.

Australia’s inflation and China’s real exchange rate, 1990–2015 (%).CPIC, Australia’s inflation; RXR, China’s real exchange rate. Source: ERS-USDA (2017).

The discussion above shows the complexity of the trend and movements of the main indicators in Australia and China. This complexity might be seen conceptually to impact dynamically on Australia’s economic growth and China’s tourism to Australia during a volatile period of increasing globalization, damaging regional and global crises and major domestic developments and reforms in the period 1990–2015. To address these important emerging and contemporary issues, and to have better, appropriate and credible insight into their potential relationships with policy relevance, we propose to study this twin relationship using a new quantitative multi-equation approach based specifically on EI theory, modelling flexibility and advanced econometric methodology, as described in the following section.

A new approach to endogenous growth and tourism modelling

Tourism economics research in past decades has been extensive and has covered many areas including approaches, methodological innovation, topics, gaps and future directions (for a comprehensive survey, see Song et al., 2012). However, with the emergence of globalization and its increasing regional EI in recent years, the international focus for growth determination has shifted to trade, FDI and tourism (services) (WTO, 2017). In spite of this important development for globalized economies, rigorous studies in general and in tourism in particular, based on its structural framework, have hardly been undertaken and reported. However, previous related studies in the specific fields of trade, official development assistance and climate change have been reported by Tran (2004, 2007) and Tran and Limskul (2013). We describe this new EI approach and methodological advances below in order to provide a new perspective and insight into tourism economics research.

The development of a simultaneous equation model for a growth-tourism causal study, and policy analysis, under an EI framework (or Systems of National Accounts 1993/2008 expenditure) is conceptually based on three theories: (a) the basic growth-determination postulates of EI and RTAs (WTO, 2017), namely trade, FDI and tourism (services); (b) gravity theory (including growth in origin and destination countries) (Frankel and Romer, 1999) and the extended Ironmonger (1972)–Lancaster (1966) new consumer demand theory where potential factors affecting tourism via its characteristics or attributes are considered and (c) Johansen (1982) add- and sub-factors, such as domestic reform and external crises, that may affect growth and tourism. They are also supported by previous successful applications as measured by modelling reliability criteria, such as that proposed by Kydland (2006) where good prediction–reality compatibility or ‘empirical fit’ is a crucial credibility criterion (e.g. Tran, 2002a, 2002c, 2004, 2005; Tran and Limskul, 2013). The economic–theoretic foundation and econometric specification and the features of the model can be briefly described as follows.

We consider, for convenience and without loss of generality, a simple model of two simultaneous (circular causality) implicit or arbitrary functions for income (Y) and tourism (T) and their key testable determinant variables. In this model, the underlying theoretical assumptions and testable hypotheses are as follows. First, Australia’s growth (Y) is determined principally by trade or openness (O) (WTO, 2017), FDI (see also Tang et al., 2007 for the possible relationship between FDI and tourism), China’s tourism (T), economic policy (W) and shocks or reforms (S) (Johansen, 1982; Tran, 2004). Second, tourism is simultaneously determined by both Australia and China’s economic demand conditions such as their growth (i.e. Y and YT, respectively; also known as the gravity factors, Frankel and Romer, 1999), destination cost of living or inflation (I), China’s real exchange rate (RXR) (Gerakis, 1965), FDI (Tang et al., 2007), W and other non-economic factors S. This model incorporates, in one important structural specification aspect, not only economic factors but also geographic or demographic attributes (Frankel and Romer, 1999; Johansen, 1982) or demographic dynamics (Kydland, 2006). Thus for simplicity and in implicit (function-free) functional form, the two functions for Y and T can be written for a sample N as

Y_{t} = F 1 (a, O_{t}, {FDI}_{t}, T_{t}, W_{t}, S_{t}) t = 1, \dots, N

T_{t} = F 2 (b, Y_{t}, Y T_{t}, I_{t}, {RXR}_{t}, {FDI}_{t}, W_{t}, S_{t}) t = 1, \dots, N

where F1 and F2 are two implicit functions linking simultaneously income and tourists to their theoretically plausible and empirically testable determinants (variables), and a and b are two vectors of parameters. In this model, Y may be defined as gross national product (GNP) or, by more popular convention GDP, or income per head of population (Easterly, 2007). T is defined as short-term arrivals (tourists) and O exports or imports or, more conventionally, openness (exports plus imports/GDP). FDI denotes foreign direct investment and S is a vector representing shocks or policy reforms. YT is the origin country’s growth representing its general economic or demand condition or supply of tourists. W denotes other economic variables (fiscal, monetary, trade and tourism policy – see Sala-i-Martin, 1991), and S represents non-economic variables (e.g. country size or population, policy reforms and external shocks – see Blake and Sinclair, 2003; Johansen, 1982; Tran, 2005; and Smeral 2009 for justification) relevant to a country’s growth and tourism policy. Importantly for our empirical study, in addition to Y, YT, O, FDI, T and S, data for W must be available and consistent with published time series data in a standard Kuznets-type accounting framework (e.g. system of national accounts, SNA93), or the accounting system of Stone (1988), or the recent World Bank tables.

As models (1) and (2) are in implicit form, they assume no specific functional form and therefore are not statistically estimable, and our purpose is ultimately to derive elasticities for their economic variables. Thus, for our empirical implementation, we use planar approximations (thus ignoring higher-order differentials) and invariant transformations (e.g. see Allen, 1960, and derivation in Tran, 1992) for (1) and (2). These two simultaneous equation models (1) and (2) in planar approximations can be written more explicitly in stochastic form and in terms of the rates of change for the continuous economic variables (denoted by y, yt, o, fdi, t, w, i, rxr and w) and binary S of all the included econometrically exogenous and endogenous variables as (for t = 1,…, N)

y_{t} = a 1 + a 2 o_{t} + a 3 {fdi}_{t} + a 4 t_{t} + a 5 w_{t} + a 6 S_{t} + u_{1 t}

t_{t} = b 1 + b 2 y_{t} + b 3 y t_{t} + b 4 i_{t} + b 5 {rxr}_{t} + b 6 {fdi}_{t} + b 7 w_{t} + b 8 S_{t} + u_{2 t}

In equations (3) and (4), y is growth (the rate of change in real GDP) and the equations are linear and interdependent or simultaneous, while a1 and b1 are constant terms, a2 to a5 and b2 to b7 are the elasticities and a6 and b8 are impact parameters. The us are other unknown factors outside the model (Frankel and Romer, 1999) or the disturbances with standard statistical properties. In models (3) and (4), circular and instantaneous causality in the sense of Granger (1969) or Engle-Granger (1987) exists or is regarded in our study as a testable hypothesis. In their exact or nonstochastic forms (in which all disturbances are idealistically zero), these equations form the basic structure of the computable general equilibrium/global trade analysis project (CGE/GTAP)) models of the Johansen class, in which all elasticities are usually assumed (calibrated) to be given or known a priori and the impact of endogenous or endogenized variables (say T) on Y is dependent on the exogenous variables and calculated system-wise, using such iterative procedures as the Gauss–Euler algorithm with a known sparse matrix of elasticities.

It can be verified that our so-called flexible (or function-free) growth and tourism equations (3) and (4) in the model above are econometrically identified in the sense of mathematical consistency. They are simultaneous equations in an endogenous growth model that require not ordinary least squares (OLS), but appropriate system estimation to produce consistent estimates. An impact study of endogenous T (or exogenous W and S) on growth can be analysed directly via its 2SLS (two-stage least squares or adjusted reduced form) or instrumental variables (IV) or by three-stage least squares (3SLS) estimation; or indirectly via its reduced-form estimation in terms of all the exogenous economic and non-economic variables in the model. Usual diagnostic tests for OLS estimation except R-squared and Durbin–Watson (DW) statistic on the estimated residuals in these IV estimation cases are not applicable. It is well known in the pure theory of econometrics that the use of OLS to estimate equation (3) or (4), for example, will, in this case, produce biased parameter estimates and subsequent incorrect policy prescriptions. Therefore, the estimation for our model below is the 3SLS that appropriately, and simultaneously, takes into account the information and effects of the two interdependent equations.

An important feature of our modelling approach adopted above is that contrary to the CGE/GTAP restrictive and so-called confirmatory approach (i.e. the causal functional relations are a priori fixed and the values of elasticities are assumed or subjectively given – see also Kydland 2006, for a requirement of data-based calibration for credible policy analysis), our impact study is historically data consistent as all required constant terms, elasticities and impact parameters are estimated from the model, and from available official data, and have asymptotically and statistically desirable and consistent properties (an important issue in empirical applications – see Frankel and Romer, 1999) when suitable estimation and forecasting methods (e.g. 2SLS or other IV methods such as the 3SLS) are employed. Another important feature is that contrary to other SNA93-based or Keynesian system-wide modelling approaches, our impact study has general flexibility in modelling the specification rationale and in implementation, assuming explicitly no a priori functional forms (e.g. linear, log, log-linear) for the equations in the model (for the relevance of this approach in preferred applied modelling, see Minier, 2007), and it can handle data on trade or budget deficits (having therefore negative values) and real rate of interest when inflation exceeds the nominal interest rate. The usual method of routine log transformations for all variables in a single or multi-equation econometric model cannot do this.

It is interesting to note that from our model’s dynamic construct (Morley, 2009), the impact may be regarded as long run in the context of Engel-Granger (1987) co-integration or long-run causality, if all variables in the equations are integrated by degree one I(1) or as short-run causality in the context of Granger (1969) causality if they are all integrated by degree zero I(0).

Empirical implementation and substantive findings

Data

Data sources

In addition to the key economic and tourism variables mentioned in the ‘Trend and pattern of Australia’s main growth indicators’ section earlier, W in the tourism equation (4) includes conventional demand – theoretically Australia’s cost of living and China’s international trade real exchange rates, and FDI (Tang et al., 2007). Data for the estimation were obtained from the ABS (2017), ADB (2017), UNCTAD (2017) and USDA-ERS (2017) databases. All economic and trade data are in real values or equivalent. In our study, all original data are obtained or derived as annual, and then transformed to their ratios (when appropriate). The ratio variables include merchandise trade and FDI. Other non-ratio variables include population (a gravity factor proxy for time series models, Frankel and Romer, 1999), inflation, real exchange rates and qualitative variables representing the occurrence of the economic, financial and other major crises, policy shift or reforms over the period 1992–2015.

Variables definition and data processing

The qualitative binary variables reflect, in a conventional manner, the major domestic, regional and global event dates, with the assumption of long-term non-decaying effects on growth and tourism. All non-binary variables are then converted to their percentage rates of change. The use of this percentage measurement (which is equivalent to log difference for small changes) is a main feature of our policy modelling and impact approach, as it deals with empirical implementation of the implicit functions (1) and (2) and avoids the problems of restrictive and potentially unsuitable a priori known functional forms (see above), and also of logarithmic transformations for negative data (such as budget (fiscal) deficits and real interest rates or current account deficits). In addition, in the model, we assume a unidirectional direction of trade and endogenous tourism to growth in a ‘causal’ context. That is, the model deals with China’s trade (in goods, FDI and endogenous tourism) and their causal impact on Australia’s growth and not vice versa. Major reforms and crises and economic variables that have been identified or assumed as exogenous or acceptable instrumental variables, affecting Australia’s growth and China’s tourism to Australia, are listed in the empirical findings table in the next section.

The p values for the weighted-symmetric unit root test for all variables in the model are given as follows: Australia’s growth = 0.230, China’s growth = 0.851, Chinese tourism = 0.083, Openness = 0.291, Openness (ROW) = 0.031, FDI/GDP = 0.070, China’s RXR = 0.042, Australian inflation = 0.927 and China’s population = 0.995. All variables used in the estimation are stationary at the 1% significance level.

The estimated model and modelling performance

To provide insights into China’s tourism, and the various key contributing factors to growth and endogenous tourism in Australia, the models (3) and (4) have been appropriately estimated, as mentioned earlier, by the 3SLS using the available data for the period 1992–2015. The basic findings are reported in Table 1. The model is identified according to the order identification tests, and all included (non-binary) variables have been found to be statistically stationary according to the usual unit root tests. The modelling performance of the estimated equations as measured by the Kydland (2006) data model compatibility or simply ‘empirical fit’ criterion and displayed graphically in Figures 4 and 5. In addition, modelling performance is measured by their empirical statistical characteristics, using Theil–MSE decomposition, and given in Table 2. As mentioned earlier, other standard diagnostic tests available for OLS estimation and residuals are not appropriate for 3SLS residuals. As assessed by these various modelling diagnostics available and reported, the estimated model first performs very well in emulating the volatile movements, peaks and troughs, especially the turning points of Australia’s growth and China’s tourism data over the sample period. Second, the Theil–MSE findings show the closeness of data, and the model first two moments, and the especially high covariance of 0.857 and 0.988 for the growth and tourism equations, respectively. The model’s residuals have also been tested for evidence of unit roots, with a p value of 0.301 for growth and 0.119 for tourism establishing statistical stationary. In addition, in the estimated model, the values for R ² (0.548 for growth and 0.616 for tourism) and DW (2.012 for growth and 2.020 for tourism) appear acceptable and show no first-order autocorrelation problem.

Table 1.

Impact of endogenous China’s tourism on Australian growth (3SLS estimates, 1992–2015).

Variables	Australian growth	China tourism
Constant	4.383***	78.480***
ROW trade/GDP	−0.032
China trade/GDP	−0.069***
FDI/GDP	−0.0001	0.017
China tourism	0.001
Post-AFC 1999	1.186**
New York terrorist attacks 2001	−1.364***
GFC 2008/2009	−1.399***	−40.730***
Australian growth		−7.060**
China’s growth		−6.823***
China population		51.503***
China real exchange rate		−1.171***
Australian inflation		1.273
Iraq War 2003		−161.989
Post-Iraq War 2004		177.088
Euro recovery 2010		15.051**
RSQ	0.548	0.616
DW statistics	2.012	2.020
ADF test on residuals	0.301	0.119

Note: 3SLS: three-stage least squares; ROW: rest of the world; GDP: gross domestic product; FDI: foreign direct investment; AFC: Asian financial crisis; GFC: global financial crisis; RSQ: R-squared; ADF: p value of the augmented Dickey–Fuller Unit Root Test. Software used for estimation is TSP-Oxmetrics6.

**Significant at the 5% level; ***significant at the 1% level.

Figure 4.

Modelling Australia’s growth, 1992–2015. YC and YC3 denote Australian growth and its 3SLS prediction. 3SLS: three-stage least squares. Note: Own calculations.

Figure 5.

Modelling China’s Tourism, 1992–2015. Note: Own calculations. TO and TO3 denote China’s tourism to Australia and its 3SLS prediction. 3SLS: three-stage least squares.

Table 2.

Friedman–Kydland modelling performance statistical characteristics of the estimated model of growth and tourism, 1992–2015 Theil–MSE decomposition.

	Australian growth		China tourism
	Data	Model	Data	Model
Mean	3.206	3.206	19.372	19.372
SD	1.011	0.754	12.080	12.993
RSQ	0.548		0.616
MSE	0.443		65.428
Mean error	0.000		0.000
Bias	0.000		0.000
Variance	0.143		0.012
Covariance	0.857		0.988

Note: RSQ: R-squared: MSE: mean-squared error. Bias + Variance + Covariance = 1 (see Pindyck and Rubinfeld, 1998).

The discussions of the findings and policy implications for China’s tourism to Australia, and its impact on Australian growth based on these empirical findings, are given in the next section.

General findings and policy implications

As mentioned earlier, the literature of tourism and its impact and contribution to economic growth since the early 1960s has been extensive with diverse empirical and simulation findings (see Song et al., 2012). However, in recent years, fast rising globalization and widespread EI (WTO, 2017) have focussed the sources of growth on international trade (or openness), FDI flows and services (in which tourism is the major component), rather than the traditional production sector of the economy. This requires new directions in research and policy analysis that better reflects these global developments.

This article makes use of this contemporary focus to develop a new approach to address these developments, the so-called EI or system of national accounts (SNA) expenditure approach (Tran, 2004; Tran, 2007; Tran and Limskul 2013), to provide substantive evidence for policy analysis in the specific case of China’s tourism, and its impact on Australia’s growth. The findings by 3SLS estimation, using 1992–2015 data of the models (3) and (4) with reported results in Table 1, show interesting results and insights for the impact of globalization, China’s tourism and regional and global crises on Australia’s growth and, importantly, the major contributing factors to China’s tourism to Australia.

It should be noted that as these findings are from an endogenous and simultaneous multi-equation econometric study with acceptable empirical fit (see above), these time series data-based findings represent another perspective of macroeconomic modelling and real-life data and may not be consistent with expectations or with other findings from alternative approaches such as input-output analysis, CGE simulation, Granger short-term causality, Engle-Granger long-term co-integration or regression analysis (see details of these approaches in Song et al., 2012).

First, during the period of mainly slow growth in Australia and declining growth in China since the GFC in 2008, the growth findings show that the age of increasing globalization with expected higher growth, non-China trade (elasticity = −0.032) and FDI (elasticity = −0.001), had no significant impact on Australia’s growth, but Australia’s trade with China (elasticity = −0.069) did, although in a negative way. An explanation for this could be that Australia, with its long history of being a quite open free-market economy, did not benefit significantly from more globalization or from increased trade and FDI which had been historically low on a relative international basis. The significant but negative impact of China’s trade on Australia is unexpected for economic analysts and policymakers, as China has been considered Australia’s biggest trade partner with significant impact on the economy in recent years. This macroeconomic effect in contrast to microeconomic expectations needs further research. However, the findings show that China’s tourism to Australia, in this environment, did have some positive macroeconomic impact (elasticity = 0.001) on Australia, although this effect is weak and statistically not significant. On the other hand, the three major crises during the sampling period 1992–2015, namely the post-AFC crisis recovery (impact = 1.186) of 1999, the terrorist attacks on New York in 2001 (impact = −1.364) and the GFC (impact = −1.399) starting in October 2008, were seen to have significantly affected Australia’s growth. One implication is that while global crises with wide-spread contagion were expected for Australia, the regional recovery after the AFC possibly reflects the importance of the closer trade and economic relationship and crisis management between Australia and its neighbouring Asian economies.

Second, the tourism findings not only recognize endogeneity in Australian growth but also provide useful and important insights into what determines or motivates China’s tourism to Australia. As a special de-commodity in the consumer demand basket with international characteristics or attributes, China’s tourism is seen as being affected by both Australia’s growth (elasticity = −7.060) and China’s growth (elasticity = −6.823). While Australia’s growth might reflect higher costs of visit to Chinese tourists, an explanation for the negative impact of China’s growth on its tourism to Australia is probably because of the diversion of Chinese tourists to other alternative attractive destinations such as Europe, as the country’s growth and therefore income rose. This is an outcome that would raise international tourism competition and be of concern to Australia’s tourism agencies and policymakers that may lead to beneficial tourism innovation in Australia.

The findings also show interestingly that, even with its restrictive population policy until recently, but with a growth rate of about 0.5% per year since the early 1990s, China’s population still exerted some contributing effects on its tourism (elasticity = 51.503) to Australia. These tourists also considered the country’s real exchange rates as a cost deterrent to travelling (elasticity = −1.171). However, the cost of living in Australia did not seem to have deterred China’s tourists (elasticity = 1.273) from coming to Australia, although the evidence is statistically weak. Additionally, it is an important finding that the impact of FDI on China’s tourism to Australia (elasticity=0.017) through, as normally speculated, investment in tourism infrastructure was found to be existent but only weak. Finally, it is interesting that major crises and reforms such as the Iraq War (impact = −161.989) and its post-war recovery (impact = 177.088) did not affect (statistically significantly) China’s tourism to Australia and tourism accelerated (impact = 15.051) after the Euro recovery that emerged in 2010 (Tran, 2002b).

Conclusion

The article addresses two important contemporary issues, namely, the contribution of Chinese tourism to Australia’s economic growth and the lack of rigorous studies taking into account the structure of modern EI theory, as applied to these two globalized trade partners. The new approach introduced in the article, which is particularly consistent with contemporary global economic and trade policy developments and modelling methodological advances, to studying what motivated China’s tourism to Australia, and whether it has had any impact on the Australian economy during the volatile period 1992–2015, has provided a number of interesting results. These results are useful for further scholarly analysis and also of policy relevance for tourism and economic policymakers. The main conclusions are the following: As part of the globalizing process, China’s tourism to Australia, while growing exponentially in the past decades and currently regarded as a critical sector to national growth, has been found to exert only a small and statistically weak contribution to the Australian economy, at least at the macroeconomic level. The economic relationships in China’s tourism to Australia have been complex and dependant on both Australia and China’s economic demand and supply conditions and international competition, as well as regional and global crises and reform developments. The findings are supported by rigorous economic–theoretic considerations and robust econometric modelling analysis.

Further research on an enlarged multi-equation EI model of endogenous growth and tourism, and extended data, would be desirable to provide further useful insights for scholarly study and for policy analysis in this important field.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

ABS (2017) Overseas Arrivals and Departures. Available at: http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/3401.0Jul%202017?OpenDocument (accessed 25 September 2017).

ADB (2017) Key Indicators for Asia and the Pacific. Available at: https://www.adb.org/publications/series/key-indicators-for-asia-and-the-pacific (accessed 25 September 2017).

Allen

RGD

(1960) Mathematical Analysis for Economists. London: Macmillan.

Blake

Sinclair

(2003) Tourism crisis management: US response to September 11. Annals of Tourism Research 30(4): 813–832.

Easterly

(2007) Was Development Assistance a Mistake? American Economic Association Meeting, 5–7 January 2007, Chicago.

Engle

Granger

CWJ

(1987) Co-integration and error correction: representation, estimation and testing. Econometrica 55: 251–276.

ERS-USDA (2017) International Macroeconomic Data Set. Available at: https://www.ers.usda.gov/data-products/international-macroeconomic-data-set/ (accessed 10 February 2017).

Frankel

Romer

(1999) Does trade cause growth? American Economic Review 89: 379–399.

Gerakis

(1965) Effects of exchange-rate devaluation and revaluation on receipts from tourism. International Monetary Fund Staff Papers 12: 365–384.

10.

Granger

CWJ

(1969) Investigating causal relations by econometric models and cross-spectral methods. Econometrica 37: 424–438.

11.

Guthrie

(1961) Demand for tourists’ goods and services in a world market. Papers and Proceedings of the Regional Science Association 7: 159–175.

12.

Ironmonger

(1972) New Commodities and Consumer Behaviour, University of Cambridge, Department of Applied Economics Monograph No. 20. Cambridge: Cambridge University Press.

13.

Johansen

(1982) Econometric models and economic planning and policy: some trends and problems. In: Hazewinkle

Rinnooy Kan

AHG

(eds) Current Developments in the Interface: Economics, Econometrics, Mathematics. Boston: Reidel, pp. 91–122.

14.

Kydland

(2006) Quantitative aggregate economics. American Economic Review 96(5): 1373–1383.

15.

Lancaster

(1966) A new approach to consumer theory. Journal of Political of Economy 84: 132–157.

16.

Minier

(2007) Nonlinearities and Robustness in Growth Regressions. American Economic Association Meeting, 5–7 January 2007, Chicago.

17.

Morley

(2009) Dynamics in the specification of tourism demand models. Tourism Economics 15(1): 23–39.

18.

Pindyck

Rubinfeld

(1998) Econometric Models and Economic Forecasts. Sydney: McGraw-Hill, 1998.

19.

Sala-I-Martin

(1991) Comment. In: Blanchard

Fischer

(eds) NBER Macroeconomic Annual 1991. Cambridge, MA: MIT Press, pp. 368–378.

20.

Smeral

(2009) The impact of the financial and economic crisis on European tourism. Journal of Travel Research 48(1): 3–13.

21.

Song

Dwyer

. (2012) Tourism economics research: a review and assessment. Annals of Tourism Research 39(3): 1663–1682.

22.

Stone

(1988) Progress in balancing the national accounts. In: Ironmonger

Perkins

JON

Tran

(eds) National Income and Economic Progress: Essays in Honour of Colin Clark. London: Macmillan, pp. 21–31.

23.

Tang

Selvanathan

(2007) The relationship between foreign direct investment and tourism: empirical evidence from China. Tourism Economics 13(1): 25–39.

24.

Tran

(1992) Modelling output growth: a new approach. Economics Letters 38: 279–284.

25.

Tran

(2002a) Korea, China and Japan’s trade with the world and its impact on new Asian regionalism ASEAN+3. Paper presented at the Joint Conference of the Association of the Korean Economic Studies, Korea Development Institute, and Research Center for International Economics, 21–22 July 2002, Yonsei University, Seoul.

26.

Tran

(2002b) Economic Crisis Management. Northampton, MA: Edward Elgar.

27.

Tran

(2002c) New Asian regionalism and ASEAN+3 free trade agreement: theoretical and empirical foundation, policy challenges and growth prospects. Chulalongkorn Journal of Economics (Thailand) 14(3): 366–384.

28.

Tran

(2004) Korea’s trade, growth of trade and the world economy in post-crisis ASEAN+3 free trade agreement: an econometric and policy analysis. Journal of the Korean Economy 5(2): 73–108.

29.

Tran

(2005) Modelling the impact of China’s WTO membership on its investment and growth: a new flexible Keynesian approach (Comment). In: Heiduk

Karyiu

Wong

(eds) WTO and World Trade. New York: Physica-Verlag, pp. 249.

30.

Tran

(2007) Causal empirics and political economy of official development assistance and development in Asia: the case of Vietnam. Journal of the Korean Economy 8(1): 91–119.

31.

Tran

Limskul

(2013) Economic impact of CO₂ emissions on Thailand’s growth and climate change mitigation policy: a modelling analysis. Economic Modelling 33: 351–358.

32.

UNCTAD (2017) Data Center. Available at: http://unctadstat.unctad.org/EN/ (accessed 10 February 2017).

33.

WTO (2017) Regional trade agreements. Available at: https://www.wto.org/english/tratop_e/region_e/region_e.htm (accessed 2 March 2017).