South Korea’s Tourism and Economic Growth with the Emergence of
Hallyu
Bachelor Thesis
June 2016
Faculty: Faculty of Economics and Business Student Name: Jillisa Basarah
Student Number: 10827765 Specialization: Economics Supervisor: Stephanie Chan
Statement of Originality
This document is written by Jillisa Alamanda Basarah who declares to take full responsibility
for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources
other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of
completion of the work, nor for the contents.
TABLE OF CONTENT
Introduction
4
Literature Review
5
Data and Model Specification
7
The Data
7
Dependent Variable
7
Independent Variable
7
Control Variables
8
The Model
8
Methodology and Empirical Results
9
Augmented Dickey-Fuller Test and Johansen’s Cointegration Test
11
Granger Causality
14
Conclusion
16
References
17
Introduction
Tourism activities are considered to be one of the major sources of economic growth. Tourism’s total contribution to the global economy in 2015 reached US$ 7.2 trillion (2015 prices), which equates to 9.8% of total GDP and supported 284 million jobs, or 1 in 11 of all jobs in the world (WTTC, 2016). Beside that, the contribution of the tourism industry to the world economy continue to grow by 2.6% in 2015 and expected to grow by 4 % in 2016. It is positive that tourism development not only stimulates the growth of the industry but also triggers overall economic growth (Lee and Chang, 2008). The rapid growth of tourism leads to the improvement of household incomes and government revenues directly and indirectly by means of multiplier effects, improving the balance of payments and provoking tourism government policies. Thus, increasing the economic growth by developing the tourism industry has been often used as an important economic development strategy worldwide.
In the field of tourism, Korea offers many entertainment and allures of their cultures for foreigners to experience. Korean government takes its tourism industry development seriously, especially after Hallyu occurs. The Hallyu impact to tourism was highly visible in 2006 where tourist arrival numbers continue to grow in a high numbered trend (Chen and Wei, 2009; Bok-Rae, 2015). The phenomena of Hallyu and the increase in tourist arrivals in Korea are believed to be closely related in achieving a greater growth of Korea’s economy. To understand this, we need to first understand Hallyu.
The term “Hallyu” was first popularized by the Chinese press in 1995 to describe the phenomena of the increasing popularity of South Korea’s culture through cultural exports that includes music (K-Pop), reality and drama TV-shows, cartoons and video games (Bok-Rae, 2015). Hallyu first appeared as a phenomenon when the K-drama exports in the late 1990s lead to a high increase in the number of tourist arrivals from China and Japan. Entering the 2000s, Hallyu quickly spread to the entire world beyond Asia with the development of internet and the growing number of social media. The early view from industrial and academic experts was very skeptical on the expansion of Hallyu towards Europe and North America where the mainstream Western culture continues to prevail. However, the interest in Korean culture within the aforementioned areas increased at a surprising rate, making Hallyu a global sensation (Bok-Rae, 2015). Kim Bok-Rae (2015) believes that as Hallyu grow bigger the increase of the world’s interest in Korean lifestyle and Korean products can enhance Korea’s economic performance through various modes, one of them through tourism.
With the presence of Hallyu, Korea realized the growth of their tourism industry and launch various tourism promotion activities to sustain the development. According to KTO statistics, incoming tourist numbers recorded per month was only around 3 million in 1995, the period before Hallyu. Whereas in 2014 tourist arrival numbers jumps as high as 14 million people per month. The
increase in tourist arrivals in Korea is staggering and the benefits from tourism in Korea was equally astounding. For example, Korean travel and tourism industry in 2014 contributed 5.9% of Korea’s total GDP, an increase of 2% compared to 2006 and directly supported 626,500 jobs (WTTC, 2015).
However, the existence and the direction of causal relationship between South Korea’s tourism and economic growth still remain a question in the academic world, making tourist-attracting policies ambiguous as a means of economic development tools (Oh, 2005; Chen and Wei 2006). Moreover, the emergence of Hallyu makes the causal relationship (tourism-led economic growth or economic-led tourism growth) investigation more intriguing than ever before. Will the increasing popularity of Korean culture phenomena (Hallyu) affects the relationship between tourism and economic growth?
Hence, this paper is aiming to identify whether there is a unidirectional or bidirectional causal relationship between tourism and economic growth in South Korea and whether this relationship changes with the emergence of Hallyu. For this purpose, quarterly data of four variables (Real GDP, Number of tourist arrivals, Real Effective Exchange Rate, and Inflation) for the period of 1975 to 2014 is observed and was tested with time series technique. The time series technique used includes the Augmented Dickey-Fuller (ADF) test to verify the stationarity of the variables, Johansen’s cointegration test to examine the existence of the long run relationship between tourism and economic growth, and Granger Causality test to determine the causal direction between tourism and economic development in South Korea.
Literature Review
The importance of the tourism industry directly results from the fact that it serves as a source of employment, private sector growth and infrastructure development for many countries (Gee, 1999). In Korea, the tourism industry in 2000 was responsible for the large foreign exchange earning amounted to 85.5% as well as the creation of 390,000 tourism jobs (KTO, 2000). In addition, tourist spending has served as an alternative form of exports, contributing to an enriched balance of payments through foreign exchange earnings in many countries (Oh, 2005). It is generally believed today that tourism development contributes positively to economic growth (Khalil, 2004; Salih, 2007; Lee and Chien, 2008; Garcia, 2014). Consequently, studies to prove whether tourism-led economic growth, be it in specific countries or in broader samples, are growing in numbers.
Most scholars believe that tourism and economic growth are cointegrated, or have the same stochastic trend (Balaguer and Cantavella-Jorda, 2002; Kim et al, 2004; Lee and Chang, 2007; Garcia, 2014). Two or more time series with stochastic trends can move together so closely over the long run that they appear to have the same trend component (Stock & Watson, 2015). For example, industrial experts suspect that Korean economic growth increases together as tourism in Korea were increasingly popular at the time when Hallyu boomed (Bok-Rae, 2015). Whether or not tourism and
economic growth moves in a common trend have been one of the important question investigated in many literatures. A study by Narayan et al. (2010) proves the existence of a cointegrating relationship between tourism and economic growth in Pacific Island Countries1 (PIC). Tourism, which is a significant component of the services sector, is one of the important instrumental in driving economic growth in these PICs. Martin (2004) also suggested that tourism and economic growth is positively cointegrated in Latin American2 countries. However, unlike Korea that view tourist spending as an
alternative form of export (Oh, 2005), Martin (2004) views tourism growth as a stimulus to consumption produced by incoming visitors. Thus, a positive tourism growth effect implies not only an increase in production and income but also an increase in market prices and exchange rate. This mean tourism can affect other economic aspects such as inflation and exchange rate. More interestingly, a study by Balaguer and Cantavella-Jorda (2002) found that a 5% of a sustained growth rate in foreign exchange earnings from tourism would imply an estimated increase of almost 1.5% domestic real income in the long run.
The possibility that economic expansion contributes to tourism growth rather than the other way around was also considered in the literature. According to the study by Lee and Chang (2007), in the long run, one directional causality relationships from tourism development to economic growth exists in OECD countries, but bi-directional and a reciprocal causal relationship exists in nonOECD countries. Since Korea has been an OECD member since 1996, one expects to find that Korea has uni-directional causation from tourism to economic growth. Nevertheless, the possibility that economic expansion is the one that contributes to tourism growth in Korea was proven true by Oh (2005): which finds that there was no long-run equilibrium relation between tourism and economic growth and that the tourism-led economic growth hypothesis does not apply to Korea. Oh’s results can be justified by the empirical analysis of Kulendran and Wilson (2000) and Shan and Wilson (2001), where they found a strong reciprocal relationship between international trade and economic growth in Australia and China. This discovery produces a notion that trade, as a form of export, is the cause of economic growth rather than tourism. Moreover, numerous studies (Marin, 1992; Shan & Sun, 1998; Xu, 1996) indicate that there is a strong correlation between international trade and economic development. Thus, combining the results of the international trade and economic growth studies mentioned before can support Oh’s findings. However, A recent study on South Korea by Chen and Wei (2009) show a contradicting result wherein a bi-directional relationship between tourism and economic growth. The different result from Oh’s research may be due to the fact that in Chen and Wei’s study, they put more variables such as real exchange rates and uncertainty factors.
1
Fiji, Tonga, The Solomon islands, and Papua New Guinea
2Nicaragua, Costa Rica, El Salvador, Colombia, Argentina, Bolivia, Brazil, Peru, Venezuela, Guatemala,
Based on the aforementioned literature, we can conclude that generally tourism promotes economic growth in the long run. However, whether the theory applies to South Korea or not is still debatable. Moreover, the presence of Hallyu in the midst of tourism-led growth investigation might generate a different result than expected. Hence, this paper will explore the relationship between tourism and economic growth in South Korea and also examine if there’s any distinction in the relationship before and after Hallyu occurred.
Data and Model Specification
1. The Data
This paper examines the relationship between tourist arrivals and economic growth in South Korea before and after Hallyu. Based on several studies (Balaguer and Cantavella-Jorda, 2010; Oh, 2005; Chen and Wei, 2009; Barro, 1996) about tourism and the determinants of economic growth, the following variables are chosen: real Gross Domestic Product (GDP) as the dependent variable and proxy of economic growth, Tourists Arrivals in Korea as the independent variable, and Korea’s real effective exchange rate, and consumer price index (CPI) growth are used as the control variable in this model. The dependent and control variables data are obtained from the Organization for Economic Co-Operation and Development (OECD) statistic website. The data for the independent variable is obtained from Korea Tourism Organization (KTO) statistic. The data used are quarterly and corresponds to a national level data from 1975 to 2014, with a total of 640 observations. The next part of this section will define all the used variables in detail.
1.1 Dependent Variable: Real Gross Domestic Product (GDP)
Given that the tourism-led growth hypothesis is about the contribution of tourism to the economic growth, Real Gross Domestic Product (GDP) is used as a proxy for economic growth. The variable real GDP is quarterly real GDP measuring the economy output adjusted to its constant price in 2010, so inflation is already taken into account. The Real GDP itself is measured by expenditure approach, which is the sum of consumption, investment, government expenditure, and net export. The value of GDP is expressed in billions of won.
1.2 Independent variable: Number of International Tourist Arrivals
International tourist arrivals into Korea was chosen as a proxy for tourism growth. There is an alternative and widely used measurement for tourism growth, which is tourism receipts. However, according to Gunduz and Hatemi-J (2005), tourism receipts to measure tourism growth creates a multicollinearity problem. International tourist arrivals by definition is the number of tourists who travel to a country other than that in which they have their usual residence, but outside their usual
environment, for a period not exceeding 12 months and whose main purpose in visiting is other than an activity remunerated from within the country visited.
Data of tourist arrivals are obtained from KTO and are provided in monthly frequency. Since this paper examines the variables in quarterly time series, the monthly data are converted into quarterly data by adding 3 months' data for each quarter. Note that the data on tourist arrivals refers to the number of arrivals, not to the number of people traveling. Thus, a person who makes several trips to a country during a given period is counted each time as a new arrival.
1.3 Control variables: Real effective exchange rate and Inflation
Several studies agreed that REER is a good determinant of economic growth since it measures specific country’s competitiveness in international trade (Tarawalie, 2010; AbuDalu et al, 2014). World Bank defines real effective exchange rate as the nominal effective exchange rate (a measure of the value of a currency against a weighted average of several foreign currencies—US dollar, Euro, and Japanese Yen) divided by a price deflator or index of costs. In this paper, REER is adjusted to consumer price index and expressed on 2010 year base. An appreciation of the real effective exchange rate is reflected by a decrease of the index and a depreciation by an increase in the index.
According to Andrez and Hernando (1997) in their study of growth and inflation in OECD countries, low inflation has a negative impact on economic growth rates, leading to significant and permanent deterioration in per capita income. The logic behind this is when inflation is high and unpredictable, businesses and households are predicted to perform poorly (Barro, 1996). Therefore, inflation had been chosen as one of economic growth factor. In this paper, inflation is measured by the consumer price index which reflects the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services that may be fixed or changed at specified intervals, in this case quarterly.
2. Model Specification
As mentioned earlier, the purpose of this paper is to analyze the relationship between Korea’s tourism growth and economic development and examine the change of relationship between the two variables (if there is any) before and after Hallyu. This paper uses quarterly data based on a total of 160 of observations in South Korea for four variables from 1975-2014. To explain the relationship mentioned before an equation in econometric terms the equations is introduced as follow:
𝑙𝑛𝑌$ =α + 𝛽' 𝑙𝑛𝑇𝐴$ + 𝛽* 𝑙𝑛𝑅𝐸𝐸𝑅$ + β3 𝑙𝑛𝐼𝑁𝐹$+ 𝑢$ (1) t = 1975Q1….2014Q4
where 1975Q1 is the first quarter of 1975; analogously for 2014 Q4. All the variables are expressed in natural logarithms so that elasticity can be interpreted. Where Y is the real GDP as the dependent variable and TA is the number of tourist arrivals as the independent variables of interest. Added into the model are the control variables, REER for Real Effective Exchange Rate, and INF for Inflation rate. The error term is represented by u which represents omitted factors left out by the deterministic part of the model.
However, the model above may be problematic due to nonstationary trends in time series data. Hence, a test to examine this problem should be conducted and will be discussed thoroughly in the next section.
Methodology and Empirical Results
This paper would like to investigate the relationship of tourism and economic growth and also examine the effect of Hallyu to the subject. Thus, the following hypotheses is presented regarding the relationship of tourism and economic growth in South Korea:
Hypothesis 1: There is a long run equilibrium relationship between tourism expansion and economic growth when control variables are included in South Korea.
Hypothesis 2: Tourism growth leads to economic growth (one-way causality: the tourism-led economic growth).
Hypothesis 3: Economic growth leads to tourism growth (one-way causality: the economic-led tourism growth)
Hypothesis 4: Tourism growth and economic growth cause each other (Reciprocal relationship between the two variables)
Cointegration and causality tests will be conducted to test the hypothesis in four different time frames. First, the test is conducted for period 1975 to 2014 to see the relationship between variables in Korea in general. Next, to see whether there are any changes before and after the Hallyu phase, the test is conducted for the period before Hallyu from 1975 to 1994, after Hallyu phase (Hallyu 1.0) from 1995 to 2014, and an alternative Hallyu starting phase (Hallyu 2.0) from 2006 until 2014. There is no concrete measurement of Hallyu nor a concrete time frame of Hallyu. Hence, the before and after Hallyu period and including the distinction between Hallyu 1.0 and 2.0 was determined according to a research paper about Hallyu phenomena by Kim Bok-Rae (2015).
Before applying the cointegration and causality test, the first step is to examine the stationarity of the variables by implementing a unit root test. The stationarity test tests the integration order of the variables. If the variable is integrated of order 1 or more this mean a past shock affects the realization of the series forever and a series has theoretically infinite variance and a time-dependent mean (White and Pettenuzo, 2010). As shown in Figure 1, a trend of a growing economy exists in Korea, so a nonstationary economic time-series data can be expected. Hence, in this paper
Figure 1. Time Trend of lnGDP and lnTA
There are two possible outcomes of the Augmented Dickey-Fuller test. The data can be found to be stationary at its levels or at its first differences. If the first outcome is shown in the ADF test, the investigation can continue with a simple regression analysis. However, if the latter is found to be true, the second step is to test for the existence of a long run relationship (cointegration) between real Gross Domestic Product (GDP), the number of tourist arrivals, real effective exchange rate, and inflation rate. Cointegration analysis provides a framework for estimation, inference, and interpretation when the variables are nonstationary. According to Engle and Granger (1984), two nonstationary variables that are integrated in similar order are cointegrated if one or more linear combinations that exist between them are stationary. Oh’s (2005) paper examines the cointegrating relationship between Tourism and economic growth in South Korea with the Engle-Granger two-stage approach (1987). However, the Engle-Granger two-stage approach is only able to test the existence of one cointegrating relationship. A merely bivariate analysis of tourism and economic growth long run relationship is not reliable since it omits important variables such as real effective exchange rate and inflation (Chen and Wei, 2009; Kim et al, 2006). Thus, a Vector Autoregression (VAR)-based cointegration test, such as the Johansen cointegration test is conducted to examine the long run relationship between variables. Johansen’s test for cointegration estimates long-run relationships using maximum likelihood procedure which tests for the number of cointegrating relationships and estimates the parameter of the cointegrating relationship.
Finally, to test whether the tourism-led growth hypothesis exists in South Korea in the four time periods mentioned before, Granger Causality method is used. Engle and Granger (1987) and Granger (1988) finds that if two time-series variables are cointegrated, then at least one-directional Granger-causation exists. Granger causation means that a variable time series is useful in forecasting another variable time series, or that variable X (tourism growth) value provides statistically significant information about future values of Y (economic growth). Granger test is created to handle pairs of
10 11 12 13 14 15 16 1975q1 1980q1 1985q1 1990q1 1995q1 2000q1 2005q1 2010q1 lngdp lnta
variables, which matches the goal of this paper on investigating the tourism-led growth hypothesis in the 4 different periods.
Augmented Dickey-Fuller test and Johansen’s Cointegration test
A unit root test is conducted to test for the order of integration of the variables or to verify their stationarity. Equation 1 may be problematic as a consequence of spurious regression phenomenon first described by Granger and Newbold (1974). If nonstationary trends in time series data is find in equation 1, the mean, variance, and autocorrelation of the series are mostly nonconstant through time. In this case, OLS estimates do not converge to constants and the standard t and F statistics will not have a limiting distribution (Phillips, 1986). Thus there is a need to investigate whether a series is stationary in levels (I(0)) or stationary in differences (I(1)) in order to apply the correct methodology and avoid spurious inferences.
Univariate analysis is important to carry out before determining the long run relationship between tourism and economic growth in South Korea. The stationarity of series was investigated by implementing the Augmented Dickey-Fuller (ADF) test as mentioned before. Because unit root test is sensitive to the presence of deterministic regressors, three models were estimated. An ADF model with a drift and time trend (𝐴𝐷𝐹'), with only drift (𝐴𝐷𝐹*) and without either drift and trend (𝐴𝐷𝐹2). The result of the ADF test for each variable on its levels is shown in Table 1 and on its first differences is presented in Table 2.
Table 1. Augmented Dickey-Fuller test on all variable levels
Statistics lnGDP lnTA lnREER lnINF Critical values 90% and 95% 𝐴𝐷𝐹' -2.448 -0.564 -2.388 -2.070 -2.576/-2.887 𝐴𝐷𝐹* -0.267 -2.126 -3.292 -3.078 -3.143/-3.444
𝐴𝐷𝐹2 3.571 3.622 -0.639 -1.704 -1.614/-1.950
Table 2. Augmented Dickey-Fuller test on all variable first differences
Statistics ∆lnGDP ∆lnTA ∆lnREER ∆lnINF Critical values 90% and 95% 𝐴𝐷𝐹' -3.978 -5.818 -5.265 -6.032 -2.576/-2.887 𝐴𝐷𝐹* -4.658 -5.764 -5.245 -6.005 -3.143/-3.444 𝐴𝐷𝐹2 -2.077 -4.349 -5.226 -5.965 -1.614/-1.950
In order to remove serial correlation in the residuals, lag period of 5 was chosen in this ADF test according to a lag order test (Oh, 2005). Table 1 and 2 shows test statistics regarding the null hypothesis of one unit root, or nonstationary, against the stationary alternative. The most general model of ADF test, which is the ADF with both time trend and a drift (𝐴𝐷𝐹'), state that the null cannot be rejected for each variable at levels as shown in table 1. On the other hand, Table 2 indicates that ∆lnGDP, ∆lnTA, ∆lnREER, and ∆lnINF reject the null hypothesis of a unit root against the alternative. Consequently, the results from the ADF test indicate that the variables are nonstationary at level and stationary at its first differences. This suggests that examination of the long run relationship should be proceed using cointegration techniques and a simple regression analysis is not reliable.
Cointegration test developed by Johansen (1988, 1991) and Johansen and Juselius (1990) was applied. The test was also suggested by Kim et al. (2006) on her paper investigating the relationship of tourism and growth in Taiwan and Balaguer and Cantavella-Jorda (2002) on Spain. Johansen Vector Autoregression (VAR)-based cointegration test uses to likelihood ratio tests, a trace test and a maximum eigenvalue test to test the number of cointegrating relationships.
The result of cointegration test can be seen in Table 3 where the test statistics of the trace tests is presented. Trace test tests for at most r cointegrating vectors against the alternative of at least r + 1 cointegrating relationships. Four lags were tried in the test for the equation, which should provide a suitable representation of the process generating the data given that we are dealing with quarterly time series (Balaguer and Cantavella-Jorda, 2002).
Table 3. Johansen maximum likelihood cointegration tests for all quarterly variables lnGDP, lnTA,
lnREER and lnCPIG Null Hypothesis r: number of cointegrating vectors 1975-2014 1975-1994 1995-2014 2006-2014 Critical values 95% Trace statistic r = 0 54.922 80.5812 38.254* 48.026 47.21 r ≤ 1 29.419* 21.678* 18.138 23.835* 29.68 r ≤ 2 10.303 5.421 8.084 7.886 15.41 r ≤ 3 0.063 0.342 0.0023 0.5862 3.76
Table 4. Johansen maximum likelihood cointegration tests for quarterly variable lnGDP and lnTA
Null Hypothesis
r: number of cointegrating vectors
1975-2014 1975-1995 1995-2014 2006-2014 Critical values
95% Trace statistic
r = 0 10.727* 3.625* 5.133* 17.508 15.41
r ≤ 1 0.048 0.491 0.626 0.743* 3.76
When the trace statistics are higher than the critical values, the null hypothesis of r cointegrating vectors against the alternative of r + 1 vectors is rejected. Table 3 shows that there is a presence of at least 1 cointegrating equations between quarterly lnGDP, lnTA, lnREER, and lnINF at the 5% level for the period 1975-2014, and similarly in before Hallyu period (1975-1995) and in Hallyu 2.0 (2006-2014). However, no proof of cointegrating relationship was found between the four variables in Hallyu 1.0 period. In Table 4 the results show that no cointegrating relationship was found in most of the periods, except in Hallyu 2.0.
Based on the test results, the long run relationship between economic and tourism growth and the combination of the four variables in the model exists. Meaning that the combination of the four variables is important for obtaining a consistent long run relationship as seen in Table 3.The existence of a consistent long run relationship between the variables can be possibly explained as follows: international tourism can contribute to the income increase in two ways, enhancing efficiency through competition between local firms and the increase in foreign exchange that tourism bring. A large proportion of a tourists’s expenditure spent on the consumption of non traded goods and services in the host country can impact the relative price of the nontraded goods and services for the domestic consumer. Moreover, the more incoming tourist into a country (South Korea), the more foreign currency exchanged into the local currency (KRW) which will increase the value and the competitiveness of the currency. It can be concluded that the existence of the link between the 4 variables are sustained by the effects that external competition (REER) and microeconomic development (INF) have on South Korean economic growth.
Table 3 shows that there is at least one cointegrating relationship between variables. Hence, a VEC model is constructed for period 1975-2014 to see the long run equilibrium relationship between the four variables:
Table 5. Parameter Estimation for lnGDP, lnTA, lnREER, and lnINF
Parameter Estimate (Normalized)
Variables Cointegrating vectors p-value lnGDP 1 . lnTA 0.462 0.112 lnREER -13.91 0.002 lnINF 1.91 0.013
The VEC regression model indicates that a 1% sustained growth rate in the number of tourist arrivals and inflation would imply a 0.46% and 1.91% increase in real GDP in the long run and a 1%
sustained depreciation in the real effective exchange rate would estimate a decrease GDP by 13.91%. The result from Table 5 confirms the result of table 4 that there is no significant relationship between tourist arrivals and real GDP from 1975-2014. Moreover, the parameter estimates show that both external competitiveness and microeconomic development in South Korea would have had significant effects on its economic growth in the last four decades.
Now, recall that in table 4 when the two control variables were excluded there are no cointegrating vector discovered between economic and tourism growth in most period except in the period of Hallyu 2.0. This interesting result suggests that the relationship between economic and tourism expansion after Hallyu 2.0 grew stronger and the two variables do not need to be supported by other control variables to be reliable anymore. It seems that as Hallyu trend spreads globally in 2006, the tourism industry in Korea has developed a strong connection with the economic growth.
Thus, the first hypothesis of this paper is confirmed in general. There is in fact, a long-run equilibrium relationship between economic growth, tourism growth and the control variables in South Korea. However, this is not the case for the years after Hallyu 1.0.
Granger Causality
Confirmation of the first hypothesis to be true implies that we can conduct the Granger Causality test to investigate the direction of causation between the economic growth and tourism. However, this paper will not be able to execute the Granger causality test for the quarterly period of 1975-1995 because there is no sign of cointegration relationship between tourism and economic growth. Moreover, since this paper is only interested in the causality relationship between tourism and economic growth, causality test of economic growth and control variables will not be included. The Causality tests between economic growth and tourism expansion can be expressed in this bivariate regressions (Oh, 2005; Kim et al, 2006; Khalil, 2004):
𝐺𝑟𝑜𝑤𝑡ℎ$ = 𝜇'+ 𝛼'? @ ?A' 𝐺𝑟𝑜𝑤𝑡ℎ$B?+ 𝛽'? @ ?A' 𝑇𝑜𝑢𝑟𝑖𝑠𝑚$B?+ 𝑒'$ (2) 𝑇𝑜𝑢𝑟𝑖𝑠𝑚$= 𝜇*+ 𝛼*? @ ?A' 𝑇𝑜𝑢𝑟𝑖𝑠𝑚$B?+ 𝛽*? @ ?A' 𝐺𝑟𝑜𝑤𝑡ℎ$B?+ 𝑒*$ (3)
Tourism and Growth represent the tourism growth (lnTA) and economic growth (lnGDP) respectively, 𝜇 is the deterministic component, and 𝑒$ is the error term of the model. Lag period (l) of
four was used in this model as recommended and tested with AIC and SBC method by Oh (2005), Balaguer and Cantavella-Jorda (2002) and Lee and Chien (2009).
The null hypothesis that Tourism does not Granger-cause Growth cannot be rejected in the cointegrated system if 𝛽''= 𝛽'*= 𝛽'2 = 𝛽'G= 0. Likewise, the null hypothesis that Growth does not Granger-cause Tourism cannot be rejected if 𝛽*' = 𝛽** = 𝛽*2= 𝛽*G= 0. Both hypotheses will be tested comparing the p-value of 𝒳* to the significance level of 0.05. Then, the test was performed to the three periods mentioned before to see if there are changes between the periods.
Table 6. Granger Causality tests
Null hypothesis 𝒳*
Growth does not cause Tourism (1975-2014) 0.023 Tourism does not cause Growth(1975-2014) 0.885 Growth does not cause Tourism (1975-1994) 0.00 Tourism does not cause Growth (1975-1994) 0.497 Growth does not cause Tourism (2006-2014) 0.135 Tourism does not cause Growth (2006-2014) 0.037
Table 6 presents the results of the Granger causality test with quarterly data for the period 1975-2014 to see the causal direction in South Korea in general, before Hallyu from 1975-1994, and Hallyu 2.0 from 2006-2014. From 1975 to 2014, the null hypothesis considering no causation of economic growth to tourism growth is rejected at the significance level of 5%, and the null hypothesis regarding economic growth causing the tourism growth is accepted at the 5% significance level. Hence, for both periods of 1975-2014 and before Hallyu only the third hypothesis (economic-led tourism growth) is accepted, while second (tourism-led economic growth) and the fourth (reciprocal relationship between tourism and economic growth) hypotheses must be rejected. This result confirms the empirical analysis conducted by Oh (2005). This economic-led tourism growth relationship implies that the rapid economic expansion in Korea turns out to attract more international travel in the long run.
However, tourism-led economic growth hypothesis is confirmed in Hallyu 2.0. This indicates that the causal relationship between tourism and economic growth changes after Hallyu when 2006 (Hallyu 2.0) was chosen as the starting point of Hallyu.
When Hallyu 2.0 occurs in 2006, the popularity of Korean culture extends not only throughout the Asian region but also across the Europe and north America. As shown in Figure 2, The number of tourist from 2006 shown to have peaked higher than ever before. Additionally, at that time tourism destination image had been largely enhanced through South Korea’s government’s destination branding activities, leading to significant growth of international tourism demand. From period Hallyu 2.0, we can say that the prediction of real GDP can be based on its own past value and the past value of tourist arrivals. This means, the increase in tourism activity from the year 2006 onwards affects income in South Korea, making tourism contribution to South Korea’s economic growth more significant than the years before.
Although generally, the economic-led tourism growth hypothesis exists in South Korea from 1975-2014, the causal relationship might change to tourism-led economic growth in the future. It seems that the impact of Hallyu to South Korea’s economy is bigger and more significant than expected.
Conclusion
The aim of this study is to examine the causal relationship between tourism and economic growth in South Korea and its changes with the presence of Hallyu. Since the variables included in the model are nonstationary and present a unit root test as proven by Augmented Dickey-Fuller test, Johansen’s cointegration test was used to examine the relationship. This methodology has allowed to attain a cointegrating relationship among the variables. These variables represent indicators of Korea’s economic growth, international tourism growth, external competitiveness, and microeconomic development within South Korea. The cointegration results provide proofs of the existence of a unique cointegrating vector. In contrast with the results by Oh (2005), the cointegration
0 500000 1000000 1500000 2000000 2500000 3000000 3500000 4000000 4500000 1975q 1 1976q 2 1977q 3 1978q 4 1980q 1 1981q 2 1982q 3 1983q 4 1985q 1 1986q 2 1987q 3 1988q 4 1990q 1 1991q 2 1992q 3 1993q 4 1995q 1 1996q 2 1997q 3 1998q 4 2000q 1 2001q 2 2002q 3 2003q 4 2005q 1 2006q 2 2007q 3 2008q 4 2010q 1 2011q 2 2012q 3 2013q 4
exists. Next, using the Granger causality test, similar with Oh’s (2005) research result, economic-led tourism growth was found in Korea. However, when the causality relationship was tested for the period after Hallyu 2.0 (2006-2014) the causality relationship change to tourism-led economic growth. This explains that after Hallyu happened, the rapid international tourism growth drives economic growth in the long run. Hallyu does not only enhance economic growth by increasing the export goods but also through tourism. In addition to this, the convergence of tourism and economic growth is sustained by including external competitiveness and microeconomic development.
Since Hallyu is an ongoing phenomena and is predicted to be the new hub of world’s popular culture, the impact of tourism on South Korean economy justifies the necessity of public intervention aimed at promoting and increasing international tourism demand as well as providing and fostering the development of tourism supply. Additionally, a cautionary on possible dangers derived from undervaluing financial support toward the efforts of entrepreneurial initiative and minimizing the significance of protecting natural and sociocultural resources should be made since the creative industry and cultural asset is the main attractiveness of Korean tourism.
Further research is encouraged in order to understand the relationship between tourism and economic growth fully. Regarding methodology issues, there are doubts that are cast on the appropriateness of model specification and the omission of important variables in previous studies. Moreover, improvement may also be done by choosing the right instruments for tourism. The more accurate the measure of tourism growth generated from economic impact data, the tests will produce more precise causal relationship.
References
AbuDalu, A. (2014). The Real Effective Exchange Rate Impact on ASEAN-5 Economic Growth. Int J Econ & Manag Sci, 03(02). doi:10.4172/2162-6359.1000174
Andres, J., & Hernando, I. (1997). Does Inflation Harm Economic Growth? Evidence for the OECD. doi:10.3386/w606
Balaguer, J., & Cantavella-Jordá, M. (2002). Tourism as a long-run economic growth factor: the Spanish case. Applied Economics, 34(7), 877-884. doi:10.1080/00036840110058923
Barro, R. (1996). Determinants of Economic Growth: A Cross-Country Empirical Study. doi:10.3386/w5698
Bok-rae, K. (2015). Past, Present and Future of Hallyu (Korean Wave). American Internation Journal of Contemporary Research, 5(5), 154-160. Retrieved from http://aijcrnet.com/
Chen, C., & Chiou-Wei, S. Z. (2009). Tourism expansion, tourism uncertainty and economic growth: New evidence from Taiwan and Korea. Tourism Management, 30(6), 812-818. doi:10.1016/j.tourman.2008.12.013
Du, D., Lew, A. A., & Ng, P. T. (2014). Tourism and Economic Growth. Journal of Travel Research, 55(4), 454-464. doi:10.1177/0047287514563167
Engle, R. E., & Granger, C. W. J. (1987). Cointegration and error-correction: representation, estimation, and testing. Econometrica, 55, 251–276.
Eugenio-Martin, J. L., Martín Morales, N., & Scarpa, R. (2004). Tourism and Economic Growth in Latin American Countries: A Panel Data Approach. SSRN Electronic Journal. doi:10.2139/ssrn.504482
Gee, C. Y. (1999). International tourism: a global perspective. Madrid, Spain: World Tourism Organization.
Gee, C. Y. (1999). International tourism: a global perspective. Madrid, Spain: World Tourism Organization.
Granger, C. W. J. (1988). Causality, cointegration and control. Journal of Economic Dynamics and Control, 12, 551–559.
Granger, C. W. J. and Newbold, P. (1974) Spurious regressions in econometrics, Journal of Econometrics, 2, 111±20.
Gunduz, L., & Hatemi-J, A. (2005). Is the tourism-led growth hypothesis valid for Turkey? Applied Economics Letters, 12, 499–504.
Johansen, S. (1988) Statistical analysis of cointegrating vectors, Journal of Economic Dynamics and Control, 12, 231±54.
Johansen, S. (1995) Likelihood-based Inference in Cointegrated Vector Autoregressive Models, Oxford University Press Incorporated, New York.
Johansen, S. and Juselius, K. (1990), Maximum likelihood estimation and inference on cointegration ± with applications to the demand for money, Oxford Bulletin of Economics and Statistics, 52, 169±210.
Katircioglu, S. T. (2009). Revisiting the tourism-led-growth hypothesis for Turkey using the bounds test and Johansen approach for cointegration. Tourism Management, 30(1), 17-20. doi:10.1016/j.tourman.2008.04.004
Khalil, S. (2004). Role of Tourism in Economic Growth: Empirical Evidence from Pakistan Economy. Pakistan Development Review, 2, 985-995.
Kim, H. J., Chen, M., & Jang, S. “. (2006). Tourism expansion and economic development: The case of Taiwan. Tourism Management, 27(5), 925-933. doi:10.1016/j.tourman.2005.05.011
Kulendran, N., & Wilson, K. (2000). Is there a relationship between international trade and international travel. Applied Economics, 32, 1001–1009.
Lee, C., & Chang, C. (2008). Tourism development and economic growth: A closer look at panels. Tourism Management, 29(1), 180-192. doi:10.1016/j.tourman.2007.02.013
Lee, C., & Chien, M. (2008). Structural breaks, tourism development, and economic growth: Evidence from Taiwan. Mathematics and Computers in Simulation, 77(4), 358-368.
Marin, D. (1992). Is the export-led hypothesis valid for industrialized countries? Review of Economics andStatistics,74,678–688.
Narayan, P. K., Narayan, S., Prasad, A., & Prasad, B. C. (2010). Tourism and economic growth: a panel data analysis for Pacific Island countries. Tourism Economics, 16(1), 169-183. doi:10.5367/000000010790872006
Oh, C. (2005). The contribution of tourism development to economic growth in the Korean economy. Tourism Management, 26(1), 39-44. doi:10.1016/j.tourman.2003.09.014
Phillips, P. C. B. (1986) Understanding spurious regression in econometrics, Journal of Econometrics,
33, 311±40.
Shan, J., & Sun, F. (1998). On the export-led growth hypothesis: The economic evidence from China. Applied Economics, 30, 1055–1065.
Shan, J., & Wilson, K. (2001). Causality between trade and tourism: Empirical evidence from China. Applied Economics Letter, 8, 279–283.
Tarawalie, A. B. (2010). Real Exchange Rate Behaviour and Economic Growth: evidence from Sierra Leone (1). Department of Economics, University of Sierra Leone.
Webel, K. (2009). J. H. Stock, M. W. Watson: Introduction to Econometrics. Stat Papers, 52(1), 251-252. doi:10.1007/s00362-009-0230-z
Xu, Z. (1996). On the causality between export growth and GDP growth: An empirical reinvestigation. Review of International Economics, 4, 172–184.
