Causality between business travel and trade volumes: Empirical evidence from Hong Kong

Tourism Management 52 (2016) 395e404

Contents lists available at ScienceDirect

Tourism Management journal homepage: www.elsevier.com/locate/tourman

Causality between business travel and trade volumes: Empirical evidence from Hong Kong Wai Hong Kan Tsui a, *, Michael Ka Yiu Fung b a b

The School of Aviation, Massey University, Palmerston North, New Zealand Aviation Policy and Research Center, The Chinese University of Hong Kong, New Territories, Hong Kong

h i g h l i g h t s
We analysed the relationship of business travel and travel volumes between Hong Kong and its three key trading partners.
A long-run equilibrium relationship between Hong Kong and the US was found.
The US showed bidirectional causality relationship between business travel and travel volumes with Hong Kong.
Business travel does Granger-cause trade volumes between Hong Kong and Mainland China, as well as between Hong Kong and Taiwan.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 November 2014 Received in revised form 12 July 2015 Accepted 18 July 2015 Available online xxx

The study employed the Engle‒Granger vector autoregressive model to investigate the causality relationship between business travel and trade volumes among Hong Kong and Mainland China, Taiwan, and the United States (US) from 2002Q1 to 2012Q4. This study presented evidence that a long-run equilibrium relationship (cointegration) between Hong Kong and the US. Additionally, the US showed bidirectional causality between the two time series variables; however, business travel does Granger-cause trade volumes for the case of Mainland China and Taiwan. The concept of the linkage between business travel and trade volumes was demonstrated in this study. Significantly, this study is expected to benefit Hong Kong policy makers by enabling better planning to attract and understand the fluctuations in international business visitors from its three key trading partners. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Business travel Trade volumes Causality Cointegration

1. Introduction Business tourism (e.g. business travel) is one of the main contributors to the economic development of a country or city, and it also concerns people travelling for purposes that are related to their work or business (e.g. Dwyer, Forsyth, Madden, & Spurr, 2000; Oh, 2005; Pablo-Romero & Molina, 2013; Swarbrooke & Horner, 2001). It is generally accepted that international business visitors may lead to increased international trade volumes as they travel to another nation or city and negotiate the sale (export) or purchase (import) of perishable and/or valuable goods from or to their countries, and then transport the goods using air transport. Hong Kong is one of the key international financial and banking

* Corresponding author. E-mail addresses: [email protected] (W.H.K. Tsui), [email protected]. hk (M.K.Y. Fung). http://dx.doi.org/10.1016/j.tourman.2015.07.010 0261-5177/© 2015 Elsevier Ltd. All rights reserved.

centres worldwide and one of the main international hub airports in the AsiaePacific region, as well as the gateway to the Pearl River Delta (PRD) in Mainland China (García-Herrero, 2011; Tsui, Balli, Gilbey, & Gow, 2014). Its prominent role in international trade and business transactions attracts millions of business travellers to Hong Kong every year for work and business purposes. According to Hong Kong Tourism Board, business visitors to Hong Kong grew from 5.45 million in 2004 to 7.60 million in 2013, equalling an average annual growth rate of 3.94%. In 2013, tourism expenditure associated with inbound tourism amounted to HK$332.05 billion (Hong Kong Tourism Board, 2013). Furthermore, international business visitors are considered to be a major source of foreign exchange and other tourism-related activities for a country (Heung & Quf, 2000). For example, Hong Kong's three major tourist sources (i.e. Mainland China, Taiwan, and the US) had a total of approximately 4.84 million business visitors travelling to Hong Kong for work or business purposes during 2013. The number of business visitors from Mainland China, Taiwan, and the US were 4.07 million,

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

Business visitors

0.38 million, and 0.39 million, respectively; the average trip expenditure and length of stay per business visitor from Mainland China, Taiwan, and the US were HK$8937 (3.4 nights), HK$5730 (2.6 nights), and HK$7300 (4.0 nights), respectively. It is worth noting that business travel develops in line with business cycles, and the main driving force is economic growth (Denstadli, 2004; Hiemstra & Wong, 2002). Often, business travellers are sent by their organisations unwillingly on overseas business trips and involved hard work with little opportunity for pleasure compared to leisure travellers (Kulendran & Wilson, 2000b). These two types of travellers (business travellers and leisure travellers) are very different in nature. However, the demand pattern of Hong Kong's international business travel, just like that of international tourism and other economic indicators is not smooth. The time series of international business travel always shows frequent fluctuations and exogenous shocks. Since 2000, the main reasons behind the fluctuations in international business visitors to and from Hong Kong consisted of the Asian financial crisis in 2001, the 9/11 terrorist attacks in 2001, the serious acute respiratory syndrome (SARS) outbreak in 2003, the global economic downturn in 2008, higher aviation fuel prices, and economic conditions in the origin countries/regions (e.g. Goh & Law, 2002; Pine & Mckercher, 2004; Song, Wong, & Chon, 2003; Song & Lin, 2009; Tsui et al., 2014). The trading activities of Hong Kong were also affected by other determinants such as seasonality and the calendar-related patterns (Goh & Law, 2002; Hiemstra & Wong, 2002). Hong Kong's total trade volumes increased from HK$4130.24 billion in 2004 to HK$7601.83 billion in 2013 (Hong Kong Census and Statistics Department, 2014a). It is evident that international trade flows become a major contributor to rapid growth in business travel (or business tourists) in Hong Kong (Hiemstra & Wong, 2002; Kulendran & Wilson, 2000b). Fig. 1 indicates that there is a close relationship between the number of international business visitors to Hong Kong and the total trade volumes, which both show similar patterns. Note that the causal relationship between international business travel and international trade has been widely studied by many academics. In general, there is a support in the literature for the idea of international travel overcoming obstacles to creating international trade between countries or regions. The Granger causality approach is one of the more popular methods to study the causality (link) between international business tourism (international business travel) and international trade. For example, Kulendran and Wilson (2000a) claimed that there was a strong long-term relationship between international

travel and international trade among Australia and its four largest trading partners (i.e. Japan, New Zealand, the United Kingdom, and the US) between 1982Q1 and 1997Q4, and also demonstrated that business travel Granger-caused total bilateral trade flows between the US and Australia. Shan and Wilson (2001) investigated the causal relationship between international trade and tourism in China using the Granger causality test. The results suggested that trade flows were linked with China's tourism for the period of 1987Q1‒1998Q1. Similarly, Khan, Toh, and Chua (2005) suggested a robust link between Singapore's business tourism and its trade volumes between 1960 and 2005, highlighting a strong link that was found between business visits and imports. Furthermore, Oh (2005) used the Engle‒Granger two-stage approach and the bivariate Vector Autoregressive (VAR) model to investigate the causality between South Korea's tourism growth and economic expansion from 1975Q1 to 2001Q1, and the findings indicated that there was no long-run equilibrium relationship between two time series variables and the tourism sector did not lead to South Korea's economic growth. In addition, Akinboade and Braimoh (2010) used the Granger causality test to show unidirectional causality running from international tourism earnings to real gross domestic product (GDP) in South Africa between 1980 and 2005, both in the shortand long-run. Our contribution to the literature emerged from a thorough investigation of the causal relationship between international business travel and trade volumes of Hong Kong and its three key trading partners (i.e. Mainland China, Taiwan, and the US). Again, it is generally accepted that international business travel and trade volumes play important roles in Hong Kong's economic development, and there are only a few empirical studies and reports in the area of Hong Kong's business travel. Therefore, we developed the econometric model to investigate whether a long-run relationship existed between business visitors and bilateral trade volumes between Hong Kong and its three main trading partners (i.e. Mainland China, Taiwan, and the US), given that these three countries took a combined share of 63.68% of Hong Kong's total international visitor numbers and 62.75% of its total trade volumes in 2013, respectively (Hong Kong Tourism Board, 2013; Hong Kong Census and Statistics Department, 2014a). Simply, a clear knowledge of the causal relationship between business visitation and the bilateral trade volumes between Hong Kong and its three key trading patterns may offer better estimation of Hong Kong's economic outlook. Another reason makes our paper meaningful relating to its contribution to increase and promote Hong Kong's international business tourism. For the strategic planning and decision-making,

8.00

9,000

7.00

8,000

6.00

7,000 6,000

5.00

5,000

4.00

4,000

3.00

3,000

2.00

2,000

1.00

1,000 0

0.00 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Trade volumes (HK$ billion)

Business visitors (million)

Fig. 1. Business visitors and trade volumes for Hong Kong (2001‒2013).

Trade volumes

396

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

it is important for policy makers (e.g. the Hong Kong government, the Hong Kong Tourism Board, airline management, and tourism operators) to understand the fluctuations in business visitors originating from the key trading partners. A clear knowledge of the fluctuations in business travellers to visit Hong Kong and the causality between business travel and trade volumes between Hong Kong and its key trading partners can help policy makers allocate resources effectively and/or design appropriate initiatives as well as recovery activities to deal with this issue strategically. This may enhance the competitiveness of Hong Kong's status as an international financial centre, the world's busiest air cargo hub, and a prominent international gateway hub to Mainland China and the AsiaePacific region. These aims could be achieved through the provision of attractive business travel packages (e.g. airfare and hotel promotions) and high-quality goods, as well as by upgrading tourism-related services and facilities (e.g. conference centres) to attract more high-end business visitors travelling to Hong Kong. These improvements in the level of customer satisfaction with goods and tourism-related services are likely to result in more repeat visits by business travellers, and also impact the local-based airlines' profitability with the transportation of more high-yield business travellers (e.g. Cathay Pacific, Dragonair, and Hong Kong Airlines). Thus, these activities will benefit the future growth of Hong Kong's economy and tourism demand. The format of this paper is structured as follows. Section 2 provides an overview of business visitors to Hong Kong from the three trading partners (i.e. Mainland China, Taiwan, and the US) and their bilateral trade volumes. Section 3 describes the methodology used to investigate the causal relationship between business travel and trade volumes. Section 4 provides a discussion of the key findings and policy implications of this study. Section 5 summaries the key findings. 2. Overview of business class travellers and trade volumes and descriptive statistics Fig. 2 displays the time plots of In(business class travellers)1 and ln(trade volumes) and Table 1 shows descriptive statistics for Hong Kong's three key trading partners (i.e. Mainland China, Taiwan, and the US) during the period of 2002Q1‒2012Q4 (Sabre Airport Data Intelligence, 2014; Hong Kong Census and Statistics Department, 2014a). Looking at the graphical analysis, the time series of the quarterly business class travellers for Hong Kong's three key trading partners exhibited different patterns and presented a larger degree of fluctuation (volatility). It is apparent that the US has a comparatively stable trend relative to Mainland China and Taiwan. The quarterly US' business class travellers visiting Hong Kong showed an upward trend during the study period, with the exception of the declining periods of 2003Q2 (2,696 business class travellers) and 2009Q1 (20,103 business class travellers). These declines were largely due to the SARS outbreak occurred during the period of November 2002‒July 2003 and the global economic downturn in 2008 that was trigged by the US sub-prime mortgage crisis in the second half of 2007 (Song & Lin, 2009; Tsui et al., 2014). However, there was an average of 22,664 US business class travellers visiting Hong Kong during the study period (see Table 1).

1 The Hong Kong Tourism Board publishes the yearly figures of business visitors to Hong Kong for regions and key countries from 2001. Such a short data period restricts robust analysis regarding the causality between business visitors and trade volumes between Hong Kong and its three key trading partners. Thus, business class travellers (air travellers who travelled with the first class and business class air tickets) are used as a proxy for analysing the causal relationship between business travel and trade volumes in this study (Sabre Airport Data Intelligence, 2014).

397

Similarly, these two adverse events also caused the significant falls in the number of business class travellers originating from Mainland China and Taiwan visiting Hong Kong. In particular, Taiwan's business class travellers were observed to have a declining trend after 2008Q1, which was largely triggered by the signing of cross-strait direct air-link agreement between Mainland China and Taiwan in April 2008. Importantly, Hong Kong started to lose its role as a convenient place for air passenger transit (particularly time-conscious business travellers) across the Taiwan Strait. Note that Mainland China is one of the largest trading partners for Taiwan (Chang, Hsu, & Lin, 2011; Lau, Lei, Fu, & Ng, 2012; Tsui et al., 2014). The declining trend returned to a more relatively stable level after 2011Q1 but the number of business class travellers visiting Hong Kong were far fewer than the pre-2008Q1 level. For example, the figures for Taiwan's business class travellers visiting Hong Kong were between 16,466 and 19,570 for the period of 2012Q1‒2012Q4. Moreover, Mainland China was Hong Kong's largest trading partner conventionally due to its historical closer economic tie and geographic proximity to Hong Kong; the figures for Mainland China's business class travellers visiting Hong Kong were 19,267‒ 160,162 across the study period. In addition, the sluggish economic development in Mainland China and Taiwan in 2012 also led to significant drops in business class traveller numbers to and from Hong Kong after 2012Q1. Mainland China's economy posted its slowest growth since 1999 in 2012: its GDP growth rate just achieved 7.6e7.9% for the period of 2012Q1‒2012Q4 (National Bureau of Statistics of China, 2014). Taiwan's economic growth rate also slowed from 4.19% in 2011 to 1.48% in 2012 due to the deterioration in external demand from Europe, Mainland China, and the US (Taiwan's National Statistics, 2014). In general, the weaker economy of a nation has a negative impact on its international business tourism and business travel demand (Beaverstock, Derudder, Faulconbridge, & Witlox, 2010). Furthermore, it should be noted that the time series of the quarterly trade volumes of Hong Kong's three key trading partners presented different upward trends, alongside the possibility of seasonal patterns during the study period. The time plots indicate that there was a distinct seasonal pattern in bilateral trading activities between Hong Kong and all of the three key trading partners (i.e. the third quarter of each year was the busiest trading period and the first quarter of each year was the least busy trading period) and also an upward trend over the study period. Therefore, all of the three time series were de-seasonalised and de-trended considering the existence in seasonality of trading activities during the analysis. The quarterly trade volumes between Hong Kong and Mainland China presented steady growth over the study period; the total bilateral trade volumes reached a remarkable value of HK$1019.31 billion in 2012Q4 (see Table 1). Similarly, Taiwan's quarterly trade volumes with Hong Kong showed an upward trend and reached the highest amount of HK$86.64 billion in 2012Q4. Regarding the US, its quarterly trade activities with Hong Kong were relatively stable throughout the study period, and also hit the record level of HK$145.55 billion in 2011Q4. Note that the global economic downturn in 2008 caused negative growth in trading activities between Hong Kong and all of its three key trading partners, albeit at different magnitudes (i.e. Taiwan's trade volumes were found to be more vulnerable to the negative impact of the global financial turmoil in 2008). In addition, the share of these three key trading partners towards Hong Kong's total quarterly trade volumes represented an average of 60.68% across the study period; they also took approximately a share of 58.04% and 63.49% of Hong Kong's total quarterly trade volumes during 2002Q1 and 2012Q4, respectively (Hong Kong Census and Statistics Department, 2014a). More importantly, their overall

398

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

Logarithm

Mainland China 14.5 14 13.5 13 12.5 12 11.5 11 10.5 10 9.5 9 2002Q1 2003Q1 2004Q1 2005Q1 2006Q1 2007Q1 2008Q1 2009Q1 2010Q1 2011Q1 2012Q1

ln(business class travellers)

ln(trade volumes)

Taiwan 13 12.5

Logarithm

12

11.5 11

10.5 10 9.5 9 8.5 2002Q1 2003Q1 2004Q1 2005Q1 2006Q1 2007Q1 2008Q1 2009Q1 2010Q1 2011Q1 2012Q1

ln(trade volumes)

ln(business class travellers)

Logarithm

The United States 12.5 12 11.5 11 10.5 10 9.5 9 8.5 8 7.5 7 2002Q1 2003Q1 2004Q1 2005Q1 2006Q1 2007Q1 2008Q1 2009Q1 2010Q1 2011Q1 2012Q1

ln(trade volumes)

ln(business class travellers)

Fig. 2. Time plots of In(business class travellers) and ln(trade volumes) for Hong Kong's key trading partners (2002Q1‒2012Q4).

Table 1 Descriptive statistics for Hong Kong's key trading partners (2002Q1‒2012Q4). Mainland China

Maximum Minimum Mean Standard deviation No. of quarters

Taiwan

The United States

Business class travellers

Trade volumes (HK$ billion)

Business class travellers

Trade volumes (HK$ billion)

Business class travellers

Trade volumes (HK$ billion)

160,162 19,267 102,513 35,290

1019.31 274.34 620.41 193.29

156,342 16,466 84,860 33,741

86.64 32.57 60.70 14.95

39,888 2,696 22,664 8,369

145.55 84.07 120.97 16.22

44

44

44

44

44

44

I(1) I(1) I(1) The US

Remarks: *, **, and *** indicate that the time series variable is significant at the 0.10, 0.05, and 0.01 significance level, respectively. ln(biz) represents log(business class travellers), ln(trade) represents log(trade volumes). ln(relative prices) represents log(relative prices). D represents first-order differencing of the time series variable.

I(1)

I(1) I(1) I(1) I(1)

Integration order

DOffices

6.411*** 6.411*** 4.773*** 6.715*** 6.646*** 6.646*** 2.111 2.227 1.203 1.236 2.532 2.574 I(1)

2.201 6068*** 5.993*** 6.396*** 1.527 6.581*** 0.598 0.945 0.377 0.197 1.657 2.898* I(1)

7.858*** 7.872*** 6.543*** 6.622*** 5.073*** 5.023***

Dln(trade) Integration order ln(relative prices) Dln(relative prices) Integration order Offices ln(trade)

1.379 1.420 1.550 1.519 1.940 2.206 I(1)

Integration order

9.009*** 9.112*** 9.934*** 10.311*** 3.693*** 10.047***

Business class travellers

Dln(biz)

Taiwan

2 The PP unit root test allows for the presence of unknown forms of autocorrelation with a structural break in the time series and conditional heteroscedasticity in the error term (Katircioglu, 2009; Phillips & Perron, 1988). 3 To determine whether the two time series variables are cointegrated, both time series need to be integrated of the same order, I(1) or greater (Engle & Granger, 1987).

Unit root tests

The cointegration (link) between two time series variables implies that a long-run equilibrium relationship exists (e.g. Granger, 1981; Khan et al., 2005; Kulendran & Wilson, 2000a; Oh, 2005). We employed the Engle‒Granger approach to examine the long-run equilibrium relationship between business class travellers and trade volumes between Hong Kong and its three key trading partners (Engle & Granger, 1987). To investigate the cointegration between the two time series variables in this study, we run the ordinary least squares (OLS) regressions for all the time series variables relating to Mainland China, Taiwan, and the US, respectively. In determining the cointegration between the two times series variables for Mainland China, Taiwan, and the US, respectively, the ADF unit root test was performed on the resulting residual series

Table 2 Unit root tests for the time series variables.

3.2. Cointegration test and Granger causality test

2.652* 2.621* 1.639 1.371 1.581 2.299

Trade volumes

To estimate the cointegration of the time series variables, all of the time series variables need to be stationary in order to avoid problems with spurious correlation. The Augmented Dicky‒Fuller (ADF) and Phillips‒Perron (PP)2 unit root tests were employed to test the stationary of the time series variables being investigated in this study (Dicky & Fuller, 1976; Phillips & Perron, 1988). Table 2 shows the results of the ADF and PP unit root tests, and the results indicate that all the time series variables are not stationary (i.e. have unit roots), except for the time series variable of Mainland China's business class travellers, which is stationary at the 0.10 significance level. Therefore, first-order differencing was applied to make the non-stationary time series variables becoming stationary. Thus, all the time series variables are regarded as cointegrated of order I(1) after first-order differencing.3

ln(biz)

Relative prices

3.1. Unit root tests

ADF PP ADF PP ADF PP

3. Methods and findings

Mainland China

Regional headquarters and regional offices

contributions towards Hong Kong's total quarterly trade volumes presented an increasing trend during the study period, showing an increase from HK$390.98 billion in 2002Q1 to HK$1248.35 billion in 2012Q4. In short, the quarterly business class travellers for Hong Kong's three key trading partners (i.e. Mainland China, Taiwan, and the US) presented different trends and seasonal patterns. The declining business class travellers of these three trading key partners to visit Hong Kong were caused by exogenous shocks, including the SARS outbreak in 2003 and the global economic downturn in 2008. Recently, fewer business class travellers from Mainland China and Taiwan to visit Hong Kong mainly resulted from their sluggish economic development in 2012 and the implementation of the direct air-link services between Mainland China and Taiwan since April 2008. In addition, trade volumes of these three key trading partners took a significant share of Hong Kong's quarterly trade volumes. Trade volumes between Hong Kong and Mainland China as well as Taiwan showed upward trends, but trade volumes between Hong Kong and the US were stable during the study period.

399

I(1)

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

400

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

3.2.2. Taiwan

Table 3 ADF tests for the hypothesis of cointegration. Business class travellers & trade volumes

Mainland China Taiwan The US

ADF tests

Cointegration

1.811 1.933 3.470**

No No Yes

InðbizÞt ¼ a0 þ

l X

a1i InðbizÞti þ

i¼1

þ GFC þ Cross Stratit agreement þ nt

InðtradeÞt ¼ b0 þ

l X i¼1

derived from the OLS regression analysis to check if they are stationary. The ADF unit root test results for the cointegration between the two time series variables are shown in Table 3. The critical values to check the stationary of the resulting residual series were adapted from MacKinnon (1991). It is seen that a long-run equilibrium relationship exists between business class travellers and trade volumes for the US with above the 0.01 significance level, but not Mainland China and Taiwan. Based upon Section 2, business class travellers and trade volumes between Hong Kong and its three key trading partners were significantly influenced by the SARS outbreak in 2003 and the global economic downturn in 2008. The relative price levels for business travellers to visit Hong Kong will become another important determinant of Hong Kong's business travel (Hiemstra & Wong, 2002; Lim & McAleer, 2001; Song et al., 2003). Furthermore, the number of foreign business establishments (regional headquarters and regional offices) set up in Hong Kong is expected to have a negative impact on business travel but increase trade volumes between Hong Kong. In addition, the signing of the Cross Strait agreement between Mainland China and Taiwan in April 2008 is expected to have an adverse effect upon the total number of business travellers passing through Hong Kong to Mainland China for work and business purposes. In the analysis of the relationship between business class travellers and trade volumes between Hong Kong and its three key trading partners, and also all the time series variables are at the cointegrated order I(1) after applying first-order differencing (see Table 2), the Vector Autoregressive (VAR) model can be established to explore the causality between business class travellers and trade volumes between Hong Kong and Mainland China, Taiwan, and the US. The bivariate [In(biz) and In(trade)] VAR regression models are shown in Equations (1)e(4):

3.2.1. Mainland China and the United States

l X

a1i InðbizÞti þ

i¼1

l X

a2i InðtradeÞti

i¼1

þ a3i ðofficesÞt þ a4i Inðrelative pricesÞt þ SARS þ GFC þ nt InðtradeÞt ¼ b0 þ

l X

(1) b1i InðtradeÞti þ

i¼1

l X

a2i InðtradeÞti

i¼1

þ a3i ðofficesÞt þ a4i Inðrelative pricesÞt þ SARS

Remarks: The ADF tests the null hypothesis of the non-stationary of the resulting residual series in comparison with the critical values of 3.337 and 3.900 at the 0.05 and 0.01 significance level (MacKinnon, 1991). ** and *** indicate the rejection of the null hypothesis at the 0.05, and 0.01 significance level, respectively.

InðbizÞt ¼ a0 þ

l X

b2i InðbizÞti

i¼1

þ b3i ðofficesÞt þ b4i Inðrelative pricesÞt þ SARS

b1i InðtradeÞti þ

l X

(3)

b2i InðbizÞti

i¼1

þ b3i ðofficesÞt þ b4i Inðrelative pricesÞt þ SARS þ GFC þ Cross Stratit agreement þ mt (4) where In(biz) and In(trade) represent the quarterly business class travellers and the quarterly trade volumes between Hong Kong and Mainland China, Taiwan, and the US in logarithm, respectively. Offices denote the number of regional headquarters and regional offices in Hong Kong established by the organisations from Mainland China, Taiwan, and the US, respectively. In(relative prices) indicate the differences between the price levels in Hong Kong and its three key trading partners in logarithm.4 SARS represents the SARS outbreak in 2003; it takes the value of 1 when the SARS outbreak happened between November 2002 and July 2003, 0 otherwise. GFC represents the global economic downturn in 2008; it takes the value of 1 when the global economic downturn happened between August 2007 and December 2008, 0 otherwise. The Cross Strait agreement represents the implementation of direct flight agreement between Mainland China and Taiwan; it takes the value of 1 at and after April 2008, 0 otherwise (this only applies to Taiwan's case). a and b are the coefficients; t denotes the time; l denotes the number of lagged variables; and nt and mt are the uncorrelated and white noise residual time series. Importantly, the cointegration of the two time series variables in this study may also indicate the presence of causality between them (i.e. at least one unidirectional Granger causality running from one time series variable to another time series variable) (Granger, 1988). However, the key drawback of the cointegration test is that this approach does not show the direction(s) of the causal relationship between the two time series variables (i.e. whether a time series variable Granger-causes another time series variable, with other factors held constant). Therefore, we continued to employ the Granger causality test to examine the directions of Granger causality between business class travellers and trade volumes for each of Hong Kong's three key trading partners. The choice of the Granger causality test over other techniques because of its favourable response to both large and small samples (Akinboade & Braimoh, 2010). In selecting the optimal lag length for the Granger causality test, we adopted the smallest values of the Akaike Information Criterion (AIC) and Schwarz Information Criterion (SIC). To examine the directions of Granger causality between the two time series variables in this study, we established the hypotheses shown in Equations (5) and (6), and also tested the null hypothesis (H0) with the conventional F-test. In performing the F-test, the null hypothesis is to be rejected when the p-value is smaller than 0.05 (or 5%); for example, a rejection of the null hypothesis in Equation (5) means that one

þ GFC þ mt (2)

4 Relative price levels take into account the Consumer Price Index (CPI) of Hong Kong and its three key trading partners and the exchange rate between Hong Kong and its three key trading partners' currencies.

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

time series variable (i.e. business class travellers ‒ In(biz)) does Granger-cause another times series variable (i.e. trade volumes ‒ In(trade)). H0: testing In(biz) does not Granger-cause In(trade) for Mainland China, Taiwan, and the US, respectively:

H0 : a21 ¼ a22 ¼ … ¼ a2l ¼ 0

(5)

Similarly, H0: testing In(trade) does not Granger-cause In(biz) originating from Mainland China, Taiwan, and the US visiting Hong Kong, respectively:

H0 : b21 ¼ b22 ¼ … ¼ b2l ¼ 0

(6)

The Granger causality test results are reported in Table 4. The results indicate that the rejection of the null hypothesis that business class travellers do not Ganger-cause trade volumes for Mainland China, Taiwan, and the US, implying that business class travellers from these three trading partners do Granger-cause their respective bilateral trade volumes with Hong Kong in this study. In addition, the null hypothesis that trade volumes do not Grangercause business class travellers was also rejected for the US, but not Mainland China and Taiwan; this suggests that the bilateral trade volumes between Hong Kong and the US do Granger-cause the number of business class travellers from the US visiting Hong Kong. 4. Discussion and policy implications This study reported that there is a long-run equilibrium relationship (cointegration) between business class travellers and trade volumes between Hong Kong and the US. Additionally, the

401

bidirectional Granger causality relationship (two-way Granger causality) between business class travellers and trade volumes was found in the case of the US, which suggested an ‘endogeneity’ or feedback effect between business travel and trade volumes between Hong Kong and the US; however, this phenomenon does not occur in the case of Mainland China and Taiwan, indicating a unidirectional (one-way Granger causality) causal effect running from business class travellers to trade volumes between Hong Kong and Mainland China as well as between Hong Kong and Taiwan. More importantly, the evidence of two-way Granger causality between business class travellers (business travel) and trade volumes in this study supports the general idea of their reciprocal link, and is consistent with the findings of other literature investigating the Granger causality between business travel and trade or economic development in different countries (Khan et al., 2005; Kulendran & Wilson, 2000a; Shan & Wilson, 2001). Although a detailed discussion of the various causality patterns is beyond the scope of this study, special consideration must be given to the case of Mainland China. Mainland China has traditionally been the largest trading partner of Hong Kong, which is a possible reason explaining why trade volumes between Mainland China and Hong Kong do not Granger-cause business class traveller numbers; it was largely due to their closer geographical proximity. Possibly, a larger amount of Mainland China's business visitors select land transport via the city of Shenzhen to reach Hong Kong for their work and business purposes, rather than using air transport. This particular situation has been in place since 2002 and reflected by the declining trend in Mainland China's business visitors travelling to Hong Kong by air, although

Table 4 Granger causality between business class travellers and trade volumes. Time series

Granger causality H0: Business class travellers do not Ganger-cause trade volumes

H0: Trade volumes do not Granger- cause business class travellers

Mainland China Taiwan The US

Rejected (0.070)* Rejected (0.005)*** Rejected (0.001)***

Failed to reject (0.229) Failed to reject (0.358) Rejected (0.041)***

Remarks: The parentheses indicate the p-values. *, **, and *** indicate that the rejection of the null hypothesis (H0) at the 0.10, 0.05, and 0.01 significance level, respectively.

4,000

4,500

3,500

3,500

11%

3,000

11%

2,500 2,000

16%

1,500 1,000

16%

16%

15% 13%

12% 13%

26%

3,000 2,500 2,000 1,500

Trade volumes

Business visitors

4,000

1,000

32%

500

500

0

0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Business class travellers (thousands) Business visitors (thousands) Trade volumes (HKS billion) Fig. 3. Business visitors, business class travellers and trade volumes between Hong Kong and Mainland China (2002‒2012). Remarks: The figures for business visitors in 2002 only included the months of JulyeDecember; the figures for business visitors in 2003 included the quarters of Q1, Q3, and Q4. % indicates the percentage of business visitors out of the total visitors from Mainland China travelling to Hong Kong.

402

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

there has been gradual growth in the bilateral trade volumes and flows between Mainland China and Hong Kong. Fig. 3 shows that around 32% and 11% of business visitors out of Mainland China's total visitor arrivals visiting Hong Kong during 2002 and 2012, respectively (Hong Kong Tourism Board, 2013). Given this, it should not be a surprise that there is no statistical Granger causality running from trade volumes to business class travellers in the case of Mainland China. As indicated in Introduction, the findings of this study have important implications for strategic planning and policy-setting by the Hong Kong government, the Hong Kong Tourism Board, airline management, and tourism operators regarding their best strategies and/or approaches attracting international business visitors from other countries visiting Hong Kong for doing business, as well as how to allocate more resources for advertising and promotion campaigns (Fig. 1 shows that the number of business visitors (or the sector of business travel) travelling to Hong Kong has shown an increasing growth rate; also, this sector will be a relatively higher value-added market for Hong Kong's economy and tourism). Importantly, tourism policy makers and operators in Hong Kong must recognise the differences in the motivations and requirements of business tourism (business visitors) compared to other types of visitors travelling to Hong Kong. A further analysis of their unique motivations and requirements will allow policy makers in developing initiatives that foster growth in Hong Kong's business tourism sector, through the provision of attractive business travel packages (e.g. airfare and hotel promotions) and high-quality foods and cuisine, as well as upgrading tourism-related services and facilities (e.g. conference centres) to attract more high-end business travellers to Hong Kong. Furthermore, it is important to pay considerable attention to the key factors that caused any decline in business class travellers5 (high-end business visitors) visiting Hong Kong from its three key trading partners in recent years. Apart from the exogenous shocks mentioned in Section 2, the fall in business travel could be largely triggered by the changing nature of business negotiation with the advances in telecommunication technology and the evolution of ecommerce. Cost-saving is one of the key aspects of successful business operations. Multinational business organisations worldwide made tremendous efforts to lower and keep their operating costs by, for example, cutting operating budgets for overseas business trips or avoiding the purchase of business class air tickets for their staff making international trips. Also, an increasing proportion of business settlements and transactions are being handled by e-business negotiation and/or e-commerce platforms such as teleconferencing (Chiu, Cheung, Hung, Chiu, & Chung, 2005; Chu, Leung, Hui, & Cheung, 2007). Such phenomena have always been the case in Hong Kong, as it acts as one of the key international commercial and financial centres ‒ the information and communications technology (ICT) is at the centre of most successful business negotiation (Fang, Worm, & Tung, 2008). According to the Hong Kong Census and Statistics Department (2014b), the effective adoption of information and communications technology (ICT) is often seen as one of the strong driving forces behind Hong Kong's economic growth. For example, approximately 14.9% and 58.3% of Hong Kong's business establishments ordered and received goods, services or information, respectively, via electronic means during 2013. In addition, the total number of regional headquarters and regional offices of foreign business establishments in Hong Kong grew from 3119 in

5 The effect of the fares for business class air tickets upon the demand of business travel of Hong Kong's three key trading partners were not considered in this study.

2002 to 3883 in 2012. The organisations from Mainland China, Taiwan, and the US set up 274, 211, and 869 regional headquarters and/or regional offices in Hong Kong during the same year, respectively (Hong Kong Census and Statistics Department, 2012). It can be inferred that if more foreign business establishments set up their regional headquarters and/or regional offices in Hong Kong, this will lead to less business travel being initiated by those organisations. Among Hong Kong's three key trading partners, it is not surprising to notice that more Mainland China's multinational corporations (MNCs) established their regional headquarters and/or regional offices in Hong Kong after 2002. The number of Mainland China's offices in Hong Kong grew from 205 offices in 2002 to 274 offices in 2012 (Hong Kong Census and Statistics Department, 2012). More importantly, many Mainland China's MNCs now prefer to operate from Hong Kong (a world city) and leverage the potential and advantages of freedom of currency flows and accessibility of global capital markets through foreign banks and international financial institutions, together with seamless electronic information transmission and exchange through internet and telecommunication; Hong Kong is one of pre-eminent international financial and banking centres offering least restrictions for international trade and business activities (García-Herrero, 2011; Shen, 2008; Short & Kim, 1999). Arguably, firms in Mainland China still face unwanted friction of information accessibility and information flow (perhaps due to the firms must agree to the Chinese government's rule of self-censoring any information the government deems inappropriate) during their business processes (Dann & Haddow, 2008; Martinsons, 2002; Zhao, Wang, & Huang, 2008). This could be one of the key determining factors affecting Hong Kong to become a preferable location for the regional headquarters and/or regional offices of Mainland China's MNCs, ranking ahead of Shanghai and Beijing6 (Lai, 2012; Zhao, Zhang, & Wang, 2004). For the policy implication, it is imperative for policy makers in Hong Kong to realise that Hong Kong's competitiveness builds on the unique combination of these core factors (particularly the freedom of information accessibility) and its sound legal and institutional framework (e.g. laissez-faire government policy), and diligent safeguard allows Hong Kong's continued dominance in serving Mainland China (Hong Kong's largest hinterland). Moreover, the intercity trading relationships between Hong Kong and other key Chinese cities should be further developed to boost Hong Kong's role as a key service provider in serving Mainland China in the region (Shen, 2008). However, the Hong Kong pro-democracy street protests in 2015 raised concerns from international community regarding Hong Kong's future political development and its prospect to maintain the role of a premium service hub facilitating international business and trade flows (García-Herrero, 2011). In fact, the evidence shows the resilience of Hong Kong's economy in terms of its remarkable capability to regain the prosperity and stability after the sovereignty changeover in 1997 and several crises (e.g. the SARS outbreak in 2003 and the global economic downturn in 2008), benefiting from Hong Kong's strong regulatory environment and its operating principal of ‘One Country, Two Systems’ (Yang, 2006; n-Berjano, 2012). Zhao, Chan, & Ramo 5. Concluding remarks Business travel is an important sector of tourism in every country (Katircioglu, 2009). The aim of this study was to

6 Beijing and Shanghai have 88% of all MNC's regional headquarters in China (Zhao et al., 2004).

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

empirically investigate the Granger causality relationship between business travel and trade volumes between Hong Kong and its three key trading partners (i.e. Mainland China, Taiwan, and the US) using the Engle‒Granger VAR model for the period of 2002Q1‒ 2012Q4. First, the empirical results revealed a long-run equilibrium relationship (cointegration) exists between Hong Kong and the US for the two times series variables. Second, a bidirectional causality relationship was found for the US, indicating that a reciprocal link exists between business class travellers (business travel) and the bilateral trade volumes with Hong Kong. However, Mainland China and Taiwan only showed the unidirectional Granger causality running from business class travellers (business travel) to trade volumes between Hong Kong for the study period. Note that the results of this study are largely consistent with prior literature regarding the causality relationship between business travel and trade or economic development in a country. The key findings of this study have a number of significant implications for strategic planning and decision-making by policy makers (e.g. the Hong Kong government, the Hong Kong Tourism Board, airline management, and tourism operators) regarding how to devise the best strategies and/or approaches to attract those high-end international business travellers (not only those from the three trading partners discussed in this study) visiting and/or return visiting (business visitors have visited Hong Kong before) Hong Kong for doing business and staying for holidays. Policy makers in Hong Kong must understand the factors that determine business tourism can be different from the factors that determine leisure travel and other types of travel (e.g. visiting friends and relatives, and leisure). It is generally accepted that a higher level of business visitor arrivals from different countries have a significant positive impact on Hong Kong's trade activities and foreign direct investments as well as local airlines' profitability, and also their trip expenditure and spending can contribute thousands of millions of dollars towards Hong Kong's tourism sector (e.g. hotel, food and beverage, luxury goods, shopping and retailing, sightseeing tours, and entertainment) and promote Hong Kong's long-standing international reputation as the “shopping paradise”. In particular, business travel has been proved to be fundamental for Hong Kong's economic development, although there is a possibility of other factors that can positively and negatively affecting Hong Kong's economy. Again, the reporting of a long-run equilibrium relationship between business travel and the bilateral trade volumes between Hong Kong and one of its three key trading partners (the US) justify the necessity of extra efforts and measures to attract more business travellers visiting Hong Kong (also include other countries worldwide); an increase in business visitation for Hong Kong will positively affect Hong Kong's economic development. Finally, due to the importance of these three key trading partners for Hong Kong's economy and tourism, this study has implied that there is a need to further investigate and estimate their respective demand for business travel to and from Hong Kong. Therefore, an in-depth study is recommended to forecast the number of business visitor arrivals from these three countries accurately using the similar approach of Kulendran and Wilson (2000b). This would further allow insight to the long-haul and short-haul business travel markets of Hong Kong. References Akinboade, O. A., & Braimoh, L. A. (2010). International tourism and economic development in South Africa: a Granger causality test. International Journal of Tourism Research, 12(2), 149e163. Beaverstock, J. V., Derudder, B., Faulconbridge, J., & Witlox, F. (2010). International business travel in the global economy. Surrey. England: Ashgate Published Limited.

403

Chang, Y. C., Hsu, C. J., & Lin, J. R. (2011). A historic move‒the opening of direct flight between Taiwan and China. Journal of Transport Geography, 19(2), 255e264. Chiu, D. K. W., Cheung, S. C., Hung, P. C. K., Chiu, S. Y. Y., & Chung, A. K. K. (2005). Developing e-negotiation support with a meta-modeling approach in a web services environment. Decision Support Systems, 40(1), 51e69. Chu, S. C., Leung, l. C., Hui, Y. V., & Cheung, W. (2007). Evolution of e-commerce web sites: a conceptual framework and a longitudinal study. Information and Management, 44(2), 154e164. Dann, G. E., & Haddow, N. (2008). Just doing business or doing just business: google, microsoft, yahoo! and the business of censuring China's internet. Journal of Business Ethics, 79, 219e234. Denstadli, J. M. (2004). Impacts of videoconferencing on business travel: the Norwegian experience. Journal of Air Transport Management, 10(6), 371e376. Dicky, D. A., & Fuller, W. A. (1976). Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association, 74(366a), 427e431. Dwyer, L. P., Forsyth, P., Madden, J., & Spurr, R. (2000). Economic impacts of inbound tourism under different assumptions regarding the macroeconomy. Current Issues in Tourism, 3(4), 325e363. Engle, R. F., & Granger, C. W. J. (1987). Cointegration and error correction: representation, estimation, and testing. Econometrica, 55(2), 251e276. Fang, T., Worm, V., & Tung, R. L. (2008). Changing success and failure factors in business negotiations with the PRC. International Business Review, 17(2), 159e169. García-Herrero, A. (2011). Hong Kong as an international banking center: present and future. Journal of the Asia Pacific Economy, 16(3), 361e371. Goh, C., & Law, R. (2002). Modeling and forecasting tourism demand for arrivals with stochastic nonstationary seasonality and intervention. Tourism Management, 23(5), 499e510. Granger, C. W. J. (1981). Some properties of time series data and their use in econometric model specification. Journal of Econometrics, 16(1), 121e130. Granger, C. W. J. (1988). Some recent development in a concept of causality. Journal of Econometrics, 39(1‒2), 199e211. Heung, V. C. S., & Quf, H. (2000). Hong Kong as a travel destination: an analysis of Japanese tourists' satisfaction levels, and the likelihood of them recommending Hong Kong to others. Journal of Travel & Tourism Marketing, 9(1‒2), 57e80. Hiemstra, S., & Wong, K. K. F. (2002). Factors affecting demand for tourism in Hong Kong. Journal of Travel and Tourism Marketing, 13(1e2), 41e60. Hong Kong Census and Statistics Department. (2012). Companies in Hong Kong representing parent companies located outside HK. Retrieved 26 October, 2014 from http://www.censtatd.gov.hk. Hong Kong Census and Statistics Department. (2014a). Total trade with ten main countries/territories. Retrieved 26 October, 2014 from http://www.censtatd.gov. hk. Hong Kong Census and Statistics Department. (2014b). Hong Kong as an information society. Retrieved 26 October, 2014 from http://www.censtatd.gov.hk. Hong Kong Tourism Board. (2013). A statistics review of Hong Kong tourism. Retrieved 26 October, 2014 from http://partnernet.hktb.com. Katircioglu, S. (2009). Tourism, trade and growth: the case of Cyprus. Applied Economics, 41(21), 2741e2750. Khan, H., Toh, R. S., & Chua, L. (2005). Tourism and trade: cointegration and Granger causality tests. Journal of Travel Research, 44(2), 171e176. Kulendran, N., & Wilson, K. (2000a). Is there a relationship between international trade and international travel? Applied Economics, 32(8), 1001e1009. Kulendran, N., & Wilson, K. (2000b). Modelling business travel. Tourism Economics, 6(1), 47e59. Lai, K. (2012). Differentiated markets: Shanghai, Beijing and Hong Kong in China's financial centre network. Urban studies (pp. 1275e1296). Lau, Y. Y., Lei, Z., Fu, X., & Ng, A. K. Y. (2012). The implications of the establishment of direct links across the Taiwan Strait on the aviation industries in Greater China. Research in Transportation Economics, 35(1), 3e12. Lim, C., & McAleer, M. (2001). Cointegration analysis of quarterly tourism demand by Hong Kong and Singapore for Australia. Applied Economics, 33(12), 1599e1619. MacKinnon, J. G. (1991). Critical values for cointegration tests. In R. F. Engel, & C. W. J. Granger (Eds.), Long run equilibrium relationships: Readings in cointegration. Oxford, England: Oxford University Press. Martinsons, M. G. (2002). Electronic commerce in China: emerging success stories. Information & Management, 39(7), 571e579. National Bureau of Statistics of China. (2014). Gross domestic product ‒ accumulated growth rate. Retrieved 26 October, 2014 from http://data.stats.gov.cn. Oh, C. O. (2005). The contribution of tourism development to economic growth in the Korean economy. Tourism Management, 26(1), 39e44. Pablo-Romero, M. del P., & Molina, J. A. (2013). Tourism and economic growth: a review of empirical literature. Tourism Management Perspectives, 8, 28e41. Phillips, P. C. B., & Perron, P. (1988). Testing for a unit root in time series regression. Biometrika, 75(2), 335e346. Pine, R., & Mckercher, B. (2004). The impacts of SARS on Hong Kong's tourism industry. International Journal of Contemporary Hospital Management, 16(2), 139e143. Shan, J., & Wilson, L. (2001). Causality between trade and tourism: empirical evidence from China. Applied Economics Letters, 8(4), 279e283. Shen, J. (2008). Hong Kong under Chinese sovereignty: economic relations with Mainland China, 1978‒2007. Eurasian Geography and Economics, 49(3), 326e340. Short, J. R., & Kim, Y. H. (1999). Globalisation and the city. Essex: Addission Wesley

404

W.H.K. Tsui, M.K.Y. Fung / Tourism Management 52 (2016) 395e404

Longman Limited. Song, H., & Lin, S. (2009). Impacts of the financial and economic crisis on tourism in Asia. Journal of Travel Research, 3, 1e15. Song, H., Wong, K. K. F., & Chon, K. K. S. (2003). Modelling and forecasting for the demand for Hong Kong tourism. International Journal of Hospitality Management, 22(4), 435e451. Swarbrooke, J., & Horner, S. (2001). Business travel and tourism. Oxford, England: Butterworth-Heinemann. Taiwan’s National Statistics. (2014). Key economic and social indicators. Retrieved 26 October, 2014 from http://eng.stat.gov.tw. Tsui, W. H. K., Balli, H. B., Gilbey, A., & Gow, H. (2014). Forecasting of Hong Kong airport's passenger throughput. Tourism Management, 42(June), 1e15. Yang, C. (2006). The Pearl River Delta and Hong Kong: an evolving crossboundary region under “one country, two systems”. Habitat International, 30(1), 61e86. n-Berjano, C. B. (2012). Industrial structural changes in Zhao, S. X., Chan, Y., & Ramo Hong Kong, China under one country, two systems framework. Chinese Geographical Science, 22(3), 302e318. Zhao, J., Wang, S., & Huang, W. V. (2008). A study of B2B e-market in China: ecommerce process perspective. Information & Management, 45(4), 242e248. Zhao, S. X. B., Zhang, L., & Wang, D. T. (2004). Determining factors of the development of a national financial center: the case of China. Geoforum, 35(5), 577e592. Sabre airport data intelligence. Retrieved 26 October, 2014 http://www.airdi.net, (2014).

Dr. Kan Wai Hong Tsui PhD is a Senior Lecturer at Massey University, teaching aviation operations, airline and airport strategies, and aviation safety management. His research covers different areas, and includes airline and airport demand forecasting, airport productivity and efficiency, and future tourism activities and trends and its relationship with air transport industry.

Prof. Michael Ka Yiu Fung PhD is a professor at the Department of Decision Sciences and Managerial Economics at the Chinese University of Hong Kong. His research spans a wide range of interests, and includes international economic, business economics, Chinese economy, aviation, and transport.