malaysia’s tourism demand: a gravity model...

ISSN: 2289-2915© Penerbit UMT

Journal of Business and Social Development Volume 6(1) 2018: 39-50

Journal of Business and Social Development Volume 6 Number 1, March 2018: 39-50

MALAYSIA’S TOURISM DEMAND: A GRAVITY MODEL APPROACH(PERMINTAAN PELANCONGAN DI MALAYSIA: PENDEKATAN MODEL GRAVITI)

MUHAMMAD HANIF OTHMAN*, NOORIHSAN MOHAMAD, GAIRUZAZMI MAT GHANI AND MUHAMMAD IRWAN ARIFFIN

Department of Economics, Kulliyyah of Economics and Management Sciences, International Islamic University Malaysia, Gombak, Malaysia.

*Corresponding author: [email protected]

Abstract: The tourism sector in Malaysia has undergone substantial growths as a result of the efforts taken by the Ministry of Tourism in policy planning and implementation. It is the government’s long-term goal to make Malaysia as the most popular tourism destination. The growth of tourism sector in Malaysia can be measured by tourist arrivals and receipts. Since this industry is able to drive Malaysian economy forward, more studies should be done in determining factors that influence demand for tourism. Therefore, this study was conducted to identify the demand factors of tourist arrivals in Malaysia and estimate their relative importance. Through understanding the determinants of Malaysia’s tourism demand, the government can design strategies and policies to enhance Malaysia’s competitiveness as a tourist destination. This study employed the bilateral tourism flows gravity model. The model includes income level of origin country and exchange rate to control for international shocks. This study also includes supply factor which is number of hotel room availability. The data used includes 149 countries, including countries with small numbers of arrivals. The bilateral tourism flows gravity model was estimated using Least Squares Dummy Variable (LSDV). The finding shows that population of country origin, country that share common border with Malaysia, income level of origin country and OIC country are major factors that increase international tourist arrivals. Since this study principally analyses the issue from a macroeconomic perspective, it is suggested that future studies re-examine the issue by using micro level data such as survey data of inbound tourists. In doing so, such findings would be more precise and informative compared to studies reliant on macro data.

Keywords: Tourism gravity model, tourist arrival, tourism sector, Ministry of Tourism, Malaysia.

Abstrak: Sektor pelancongan di Malaysia telah mengalami pertumbuhan yang besar hasil daripada perancangan dan pelaksanaan dasar oleh Kementerian Pelancongan Malaysia. Ini juga salah satu matlamat jangka panjang kerajaan untuk menjadikan Malaysia sebagai destinasi pelancongan paling popular. Pertumbuhan sektor pelancongan di Malaysia boleh diukur dari segi ketibaan pelancong dan pendapatan hasil daripada sektor pelancongan. Memandangkan industri ini dapat memacu ekonomi Malaysia ke hadapan, lebih banyak kajian harus dilakukan dalam menentukan faktor-faktor yang mempengaruhi permintaan untuk pelancongan. Oleh itu, kajian ini dijalankan untuk mengenal pasti faktor-faktor ketibaan pelancong di Malaysia dan menganggarkan kepentingan mereka. Dengan memahami faktor penentu permintaan pelancongan Malaysia, kerajaan dapat merangka strategi dan dasar untuk meningkatkan daya saing Malaysia sebagai sebuah destinasi pelancongan. Kajian ini menggunakan model graviti dengan memasukkan faktor seperti kadar pendapatan negara asal, nilai pertukaran dan ketersediaan bilik hotel. Data yang digunakan merangkumi 149 buah negara termasuk negara dengan bilangan ketibaan yang kecil. Model graviti dianggarkan menggunakan Least Squares Dummy Variable (LSDV). Hasil kajian menunjukkan bahawa populasi negara asal, negara yang berkongsi sempadan yang sama dengan Malaysia, kadar pendapatan negara asal dan negara OIC adalah faktor utama yang meningkatkan ketibaan pelancong antarabangsa. Oleh kerana kajian ini secara asasnya menganalisis isu ini dari perspektif makroekonomi, adalah dicadangkan kajian masa depan mengkaji semula isu tersebut dengan menggunakan data tahap mikro seperti data tinjauan pelancong masuk. Dengan berbuat demikian, penemuan akan lebih tepat dan bermaklumat berbanding dengan kajian yang bergantung kepada data makro.

Kata kunci: Model graviti pelancongan, ketibaan pelancong, sektor pelancongan, Kementerian Pelancongan Malaysia.

076.indd 39 3/27/18 3:07 PM

Muhammad Hanif Othman et al. 40


IntroductionTourism industry can be classified as one of the services sector, and today this industry has become an important industry and has grown to be among the largest industry after automobiles and oil in terms of revenue (Hanafiah & Harun, 2010). According to WTO, the amount of international tourist arrivals worldwide increased from 803 million in 2005 to 1,184 million in 2015, an increase of 47%. Furthermore, the receipts from worldwide tourists in 2015 amounted to US $1,260 billion, 86% increase since 2005 (UNWTO, 2016). These figures indicate a remarkable contribution of international tourism to the global economy. Therefore, many countries around the globe began to give serious attention to the development of this sector and offer a variety of packages and promotions to attract tourists around the world and yield profits for the country (Kusni et al., 2013).

In the case of Malaysia, travel and tourism sectors total contribution to GDP in 2013 was RM158.2 billion or 16.1% and with this significant contribution, Malaysia has been ranked third among the Asia Pacific countries after Cambodia and Thailand (WTTC, 2014). Furthermore, the amount of international

tourist arrivals and receipts increased from 10.33 million and RM17.3 billion in 2000 to 25.7 million and RM69.1 billion in 2015 respectively. Despite this remarkable increase in tourist arrivals and receipts, there is perhaps a big question mark on the factors that may potentially affect international tourist arrivals to this country. Hence the objective of this article is to identify the factors and estimate their importance in explaining tourist arrivals to Malaysia. One factor raised here is the geographical imbalance in international tourism flows in Malaysia. In 2015, the ASEAN region contributed a 74.4% share with 19.1 million tourists, while the portion of the medium-haul markets and long-haul market contributed 18.8% or 4.8 million tourists and 6.7% or 1.7 million tourists to Malaysia’s total arrivals in 2015 (Malaysia Tourism Promotion Board, 2016).

Furthermore, in terms of tourist arrivals to Malaysia from Muslim and non-Muslim countries, it shows that arrivals from non-Muslim countries have been greater than those from Muslim countries. Table 1 shows the top ten countries of Muslim and non-Muslim tourist arrivals. It shows that in total, there were 4,349,721 arrivals (17% of total arrivals) from Muslim countries while 20,792,998 arrivals

1 Recently there have been many government initiatives such as halal tourism, Muslim friendly tourism, Shariah-compliant tourism to attract middle east especially Muslim countries tourists. The question is whether that initiatives or resources really contribute effectively to tourism Malaysia.

Table 1: Top ten tourist arrivals: non-OIC vs. OIC countries, 2012Non-OIC countries Arrivals Market

Share (%) OIC countries Arrivals Market Share (%)

Singapore 13,014,268 62.59% Indonesia 2,382,606 54.78%China 1,557,960 7.49% Brunei 1,258,070 28.92%Thailand 1,263,024 6.07% Iran 127,404 2.93%India 691,271 3.32% Saudi Arabia 102,365 2.35%Philippines 508,744 2.45% Bangladesh 86,465 1.99%Australia 507,948 2.44% Pakistan 79,989 1.84%Japan 470,008 2.26% Oman 24,977 0.57%United Kingdom 402,207 1.93% Kuwait 22,759 0.52%South Korea 283,977 1.37% Iraq 21,939 0.50%Taiwan 242,519 1.17% Kazakhstan 20,188 0.46%Total non-OIC 20,792,998 100% Total OIC 4,349,721 100%

Source: World Tourism Organization

076.indd 40 3/27/18 3:07 PM

MALAYSIA’S TOURISM DEMAND: A GRAVITY MODEL APPROACH 41


(83% of total arrivals) come from non-Muslim countries in 2012. More than half of the total non-Muslim arrivals (62.59%) were from Singapore, leaving 7 million arrivals from other non-Muslim countries. On the other hand, Indonesia and Brunei are the top two Muslim countries arrivals with 54.78% and 28.92%, respectively. Only about 709,045 arrivals were from Muslim countries when exclude Indonesia and Brunei. From this figure, it shows that Muslim countries may not be important sources of tourist arrivals.1 However, a formal test is needed to determine their importance in explaining tourist arrivals to Malaysia.

This study employed the bilateral tourism flows gravity model to identify the factors and estimate their importance in explaining tourist arrivals to Malaysia. Since this paper is in the context of international tourism research, the influential factors of international tourism such as exchange rate and GDP that explain about the tourism flows could not be pulled out. The gravity model used in this study differs from those previous studies such as Mohebi & Rahim (2010); Hanafiah & Harun (2010); and Kosnan, Ismail & Kaliappan (2013) in examining Malaysia’s tourism sector because the number of countries included was limited. In this study, the data includes 149 countries including countries with small numbers of arrivals since excluding them could bias the results. The bilateral tourism flows gravity model was estimated using Ordinary Least Squares (OLS), Fixed Effect Model (FEM) and Random Effect Model (REM).

The remainder of this paper is structured as follows; Section 2 discusses on the major market share of tourist arrivals to Malaysia. Section 3 provides a discussion on the theoretical framework. Section 4 explains the data and methodology used, followed by Section 5 which presents the empirical results and discussion. Finally, Section 6 concludes the paper.

Market Share of Tourist Arrivals to MalaysiaThe major markets of Malaysia tourism are shown in Table 2. It shows that 52.80% of the total international tourist arrivals in Malaysia are Singapore, followed by Indonesia, China, Thailand and Brunei with 9.67%, 6.32%, 5.12% and 5.1% respectively. There are 5 Asean countries among the top 10 countries of tourist arrivals accounted for 74.76% in 2012. Figure 1 presents the relationship of international tourist arrivals and distance between Malaysia and the origin country of the tourist. The distance is divided into 4 groups with 5 thousand kilometers interval.

The size of bubbles in Figure 1 represents the volume of tourist arrivals to Malaysia. As can been seen in this figure, Singapore has the largest bubble and it means that Singapore has the highest number of arrivals to Malaysia. The majority (89.66%) of tourist arrivals to Malaysia are concentrated from countries within 5 thousand kilometers distance from Malaysia. The next distance ranges from 5 to 10 thousand kilometers distance accounts for around 5.8% of the market share while 4.4% was accounted to countries within 10 to 15 thousand kilometers. The last distance (15 to 20 thousand kilometers) contributes very small market share which is 0.14%. The countries in these areas include Mexico, Uruguay, Brazil, Argentina and Chile, which have a large range from Malaysia. This is consistent with the concept of gravity model which fewer tourists would like to visit countries within longer distance. Tourists prefer to visit countries within shorter distance because they can reduce their cost especially the transportation costs.

Gravity ModelThe gravity model which originates from Newton’s law of gravity is the most common formulation of the spatial interaction method

1 Recently there have been many government initiatives such as halal tourism, Muslim friendly tourism, Shariah-compliant tourism to attract middle east especially Muslim countries tourists. The question is whether that initiatives or resources really contribute effectively to Malaysia tourism.

076.indd 41 3/27/18 3:07 PM



because it is “understandable and practical to measure the relationship of one zone to another zone” (Moorthy, 2014, p.261). This model has also been widely used in explaining the capital flows, volume of trade and migration of people among the countries in the world. For example, this model has been applied to estimate the effects of different factors on international flows such as the effect of migration (Gil-Pareja, Llorca & Martínez, 2007; Karemera, Oguledo & Davis, 2000), goods (Anderson & Van Wincoop, 2003; McCallum, 1995; Rose, 2000) and foreign direct investment (Eichengreen & Tong, 2007; Head & Ries, 2008).

Tinbergen (1962) claimed that the gravity model is the most suitable empirical tool to understand trade and other economic flows in the world of economy. The general gravity model that was proposed by Tinbergen can be expressed in the notation as follows:

Fij = A. (Yi x Yj)

α (1)

whereFij : is the flow “trade volume” between

country origin and destination,A : is proportionality constant,Yi x Yj : are economic sizes of country, i and j

(GNP, GDP or per capita GDP),Dij : is the distance between countries.

Table 2: Market share of major international tourist arrivals to Malaysia, 2012

From Tourist Arrivals Market Share (%)Singapura 13,014,268 52.80%Indonesia 2,382,606 9.67%China 1,557,960 6.32%Thailand 1,263,024 5.12%Brunei 1,258,070 5.10%India 691,271 2.80%Philippines 508,744 2.06%Australia 507,948 2.06%Japan 470,008 1.91%United Kingdom 402,207 1.63%Republic of Korea 283,977 1.15%United States of America 240,134 0.97%Vietnam 211,008 0.86%France 136,172 0.55%Germany 131,277 0.53%Islamic Republic of Iran 127,404 0.52%Nepal 123,173 0.50%Saudi Arabia 102,365 0.42%Netherlands 88,404 0.36%Canada 86,931 0.35%Bangladesh 86,465 0.35%Pakistan 79,989 0.32%New Zealand 65,726 0.27%

Source: World Tourism Organization, 2012

Dijγ

076.indd 42 3/27/18 3:07 PM



Figure 1: Distance and market share of international countries to Malaysia, 2012Source: World Tourism Organization, 2012

Equation (1) is the core gravity model equation of international trade as trade volume to be a function of economic size and distance, where trade volume is predicted to be a positive function of economic size and negative function of distance. The population size for both exporter and importer countries are often included as variables to predict trade flows. After including population size in Equation 1, the equation transformes as follows:

Fij=A. (Pi x Yi)

α (Pj x Yj)β

(2)

After a simple arrangement, equation 2 can be written as follows:

Fij=A. (Pi x Yi)

α (Pj x Yj)β

(3)

Equation 3 can become linear if the both side logarithms are taken in place:

logFij = A* + α log(Pi x Yi) + β log(Pj x Yj)

− γ logDij+ εij (4)

Where A* is log A, and α, β and γ are parameters to be estimated. While the random factors that affect bilateral trade is being presented by εij, also known as white noise error term with constant variance and zero mean. Now

equation 4 represents a function of per capita GDP and the distance between two countries. However, various importer countries presented by (Pi x Yi) can be dropped from the equation 4 since it cannot be a source of explanation for trade deviations (Bos & van de Laar, 2004). Therefore, the estimable model can be written as:

logFij = A* + α log(Pj) + β log(Yj)

− γ logDij+ εij (5)

In recent years, the gravity models have been repetitively used to make analysis of economic phenomena related to the flow of goods and services (Bos & van de Laar, 2004).In this respect, Morley, Rossello & Santa-Gallego (2014) have provided theoretical foundation for tourism flows that derived from individual utility maximization model. Studies that used the bilateral tourism flows gravity model include Ghani (2016); Eilat & Einav (2004); Kosnan, Ismail & Kaliappan (2013); Mohebi and Rahim (2010); Hanafiah & Harun (2010); Gil-Pareja, Llorca-Vivero & Martinez-Serrano (2007).

Empirical ModelsThis study uses the bilateral tourism flows gravity model to identify the factors and estimate

Dijγ

Dijγ

076.indd 43 3/27/18 3:07 PM



their importance, in explaining tourist arrivals to Malaysia. The gravity model has been widely used in explaining the volume of trade, capital flows, regional science economic geography as well as migration of people among countries in the world (Kosnan, Ismail & Kaliappan, 2013). In estimating tourism demand, Rodrigue (2004) uses the gravity model by Tinbergen (1962). Some adjustments are made by the Rodrigue (2004) to suit the tourism model. The model that has been proposed by Rodrigue is as follows:

TDij= K mi . mj

(6)

whereTDij : represents tourist arrivals from country

i to destination country j,K : is a constant,mi : is a factor that generate the movement

of international tourism,mj : is a factor that attracts the movement of

international tourism,Dij : is the distance between origin country,

i and destination country, j. In the case of Malaysia, studies using

the bilateral tourism flows gravity model to examine the determinants of tourist arrivals have included Ghani (2016), Kosnan, Ismail and Kaliappan (2013), Mohebi & Rahim (2010) and Hanafiah & Harun (2010), however the number of countries included in these studies was limited except study by Ghani (2016). Mohebi and Rahin (2010) use panel data set for 14 origin countries (Hong Kong, Philippine, Thailand, Saudi Arabia, Singapore, Brunei Darussalam, China, Taiwan, Indonesia, Japan, United Kingdom, Australia, Germany and United States) from 1998 to 2009 with 154 observations. Another study by Hanafiah & Harun (2010) use “a cross-sectional pool time-series of tourist arrivals from Australia, Hong Kong, Indonesia, United Kingdom, Thailand, Taiwan and China” to estimate tourism demand in Malaysia (p.200). On the other hand, Ghani (2016) includes 171 countries for year 2012 to

examine the determinants of tourist arrivals to Malaysia. However, the variables that are often used in the gravity model such as exchange rate, language and common colonizer are not used by Ghani since his main focus is the effect of Muslim country on the number of tourist arrivals. Furthermore, his study uses cross-section data rather than panel data model. Therefore, this study attempts to use panel data and include more variables that are often used in the gravity model in explaining tourist arrivals to Malaysia.

Variables and Model SpecificationThe advantage of using this gravity model is that this model can estimate both time variant and time invariant variables (Kosnan, Ismail & Kaliappan, 2013). Additionally, the model allows more factors to be included to explain the international tourist arrivals. The choice of variables used in this study depends on the availability of the data and also the objective of this study. The number of tourist arrivals have been used as a proxy of demand for tourism by the majority of researchers (Witt and Martin, 1987; Kusni et al., 2013; Mohd Salleh et al., 2007 and Li, 2004). This is due to the data availability, as data for tourist expenditures, tourism receipts and number of night are only available for a small number of countries and time periods. Study by Mohd Salleh et al. (2007) found that majority of studies used total tourist arrivals as a proxy for tourism demand. Furthermore, Tang (2014) who has reviewed 61 articles on tourism demand since 1970, found that 47 of them used tourist arrivals as dependent variable. Alternatively, some studies (e.g. Zaman, Khan & Ahmad, 2011; Song et al., 2010; Algieri, 2006; Qiu & Zhang, 1995; Lee, Var & Blaine, 1996; and Seddighi & Shearing, 1997) used the tourism expenditure as proxy for international tourism demand.2 However, only a few studies used the number of night (9%) to measure international tourism demand. Due to availability of the data, this study used tourist arrivals as indicator for dependent variable.

Dijγ

2 Tourism receipts or expenditure are the most appropriate variable to be used as a dependent variable from the perspective of the destination country (Proenca and Soukiazis, 2005).

076.indd 44 3/27/18 3:07 PM



These are similar to the studies by Mohd Salleh et al. (2007), Hanafiah & Harun (2010), Tang (2014) and Ghani (2016) who asserted that majority of tourism demand studies used tourist arrivals to measure international demand for tourism.

Numerous factors influencing tourism arrivals are highlighted in the literature. The demand for tourism is based on demand theory. Therefore, income and tourism price are the factors that influence the demand (Mohd Salleh et al., 2007). Munoz & Amaral (2000) stress that more residents can afford to travel to outside countries if a country’s income increases. Therefore, the tourist arrivals will be the positive function of income. This hypothesis is supported by many studies including Kosnan, Ismail & Kaliappan (2013); Hanafiah & Harun (2010); Mohd Salleh et al. (2008) and Zhang & Jensen (2005). Besides income level, prices are also identified as one of the important factors that determine tourism demand (Dritsakis, 2004; Munoz & Amaral, 2000). Tourism price is expected to have a negative relationship with tourism since when the cost of living in a destination country is lower than the origin country, international tourism will increase. Exchange rate is also included as the determinants of tourist arrivals (Hanafiah & Harun, 2010; Mohd Salleh et al., 2008). The nominal exchange rate measures the effective prices of goods and services in Malaysia. Any depreciation in tourist currency may discourage people to travel.

Population size was expected to have a positive relationship with tourist arrivals, as countries with lower populations are expected to have smaller number of tourists (Ghani, 2016; Fourie et al., 2015). Furthermore, distance was

expected to be negatively related with tourist arrivals since more tourists are likely to visit countries within a shorter distance to reduce the cost especially the transportation cost (Ghani, 2016). Other studies also include language and geographical borders as the determinants of tourist arrivals (Kosnan, Ismail & Kaliappan, 2013; Khadaroo & Seetanah, 2008). This would lead to increase in tourist arrivals since common border and language sharing would allow frequent tourist visits to the destination countries.

Furthermore, micro variable such as accommodation capacity is also important to examine tourist behaviour in order to formulate policies to enhance the tourism sector. To proxy accommodation capacity, the number of hotel rooms available in Malaysia is used. The number of hotel rooms available and tourist arrivals are expected to be in positive relationship. More rooms available to tourists who come to Malaysia will result in more tourist arrivals being received.

Muslim-majority countries are expected to have a positive relationship with tourist arrivals in Malaysia since Muslim is the majority population in Malaysia (Ghani, 2016). Thus, to determine the Muslim countries, this study used membership in the OIC as a proxy.

Based on the variables discussed, a bilateral tourism flows gravity models were estimated:

Arrivalijt = f (Distanceij, Populationjt, GDPCAPITAjt, Exchange Rateit, Room Hotelit, Allocationjt, OIC, Borderij,) (7)

The econometric model on estimating tourism demand used in a log-linear functional as follows:

(8)

076.indd 45 3/27/18 3:07 PM



where the dependent variable in the natural log of the number of tourist arrivals in Malaysia. this variable is denoted by Arrivalsijt. The independent variables are defined in Table 4. In the equation 8, the variables were expressed in logarithm form. β0 is the constant term, βn are the coefficients of each variable taken into consideration while εijt is the error term.

DataMost of the international tourism studies use time series data including Zaman, Khan and Ahmad (2011); Song et al. (2010); Wang (2010); Mohd Salleh, Cheah & Othman (2011), while some have used cross-sectional data (Ghani, 2016; Var, Mohammad & Icoz, 1990) and pooled/panel data (Massidda and Etzo,

2012; Kusni, Kadir and Nayan, 2013; Tang, 2014; Hanafiah & Harun, 2010; Mohd Salleh et al., 2010). Generally, the gravity model operated on the basis of cross-sectional data and panel data. However, among those types of data, panel data approach is better since the cross-sectional data has a limitation since this approach limits the analysis to a single time period. Panel data approach can capture “the relevant relationships over time and avoiding the risk of choosing an unrepresentative year” (Saray & Karagöz, 2010, p.40). Moreover, this application of panel data can help control for heterogeneity amongst countries. Therefore, it is possible to employ three basic methods in panel data approach which are OLS, Random Effects Model (REM) and Fixed Effects Model (FEM).

Table 3: Descriptive statistics for panel data, 149 countries

Variables Mean Standard Deviations Minimum Maximum

LNARRIVAL 9.10 2.58 0.69 16.41LNGDPORIGINCAP 8.74 1.67 5.23 11.36LNDISTANCE 8.85 0.76 6.23 9.88LNPOPULATION 16.50 1.86 10.84 21.02LNALLOCATION 19.50 0.58 18.49 20.23LNEXCHANGERATE 1.21 0.12 0.92 1.367LNROOM HOTEL 11.93 0.24 12.18 11.24BORDER 0.06 0.23 0.00 1.00OIC 0.28 0.45 0.00 1.00

Table 4: Definition of independent variables

Variables Description of Variable Expected SignGDPCAPITAjt : is the GDP of the tourist’s country of origin. +

Distanceij: is the geographical distance in nautical miles between country j’s capital city and i’s Kuala Lumpur, Malaysia’s capital city; -

Populationjt : is the population size of the tourists’ country of origin; +

Hotel Roomit: is the number of hotel room available in Malaysia as a proxy for accommodation capacity. +

Exchange rateit : is the exchange rate between Malaysia RM and US Dollar. -

OIC: is a dummy indicating whether the tourists’ country of origin is a Muslim country. It took a value of 1 when the country of origin is a member of OIC and 0 otherwise.

+

Borderij: is a dummy variable that assumes the value of one when the countries have a common border. +

Allocation : is the amount of government allocation in Ringgit Malaysia (RM) for the tourism industry in Malaysia, j, over time, t. +

076.indd 46 3/27/18 3:07 PM



The data for tourist arrivals from 149 countries are extracted from Tourism Malaysia. The data spans from 1995 until 2012. Therefore, the present study consists of 149 pairs of unbalanced panel data with 1,267 observations. The data includes the countries with small number of arrivals to avoid the bias results. Data for Gross Domestic Product (GDP), population and exchange rate are compiled from the World Bank’s World Development Indicators (WDI). Meanwhile, the data for variables such as distance, common border and common language are extracted from the Centre d’Etudes Prospectives et d’Informations Internationales (CEPII). For the indicator of Muslim countries, this study used membership in the Organization of Islamic Countries (OIC)3.

ResultsTable 5 presents the regression results of Ordinary Least Squares (OLS), Fixed Effect Model (FEM) or Least Squares Dummy Variable (LSDV) and Random Effect Model (REM). This study employed the bilateral tourism flows gravity model to identify the factors and estimate their importance in explaining tourist arrivals to

Malaysia. The result is based on two ways fixed effect which are fixed origin country and years.

The coefficient of distance (lndist) is positive and statistically significant. This finding is different with the results of Hanifah and Harun (2010), Kosnan et al. (2013), Deluna & Jeon (2014) and Ghani (2016). Population and neighboring countries turned out significantly increasing tourist inbound to Malaysia. These results show that an increase in population will increase tourist arrivals by 1.5%, while increase in countries that share common borders with the destination country will increase tourist arrivals since the cost of visiting destination country are relatively lower.

On the other hand, the estimated elasticity of tourist arrivals to GDP per capita of the tourist’s country of origin is 0.89, indicating that a 1% increase in GDP per capita will increase tourist arrivals by 0.89%. This finding shows that an increase in the income of tourists will increase international tourism in the destination country.

Finally, the coefficient of OIC (Muslim countries) is positive and statistically significant. This is because of the Muslim-friendly products and activities being implemented by government

Table 5: Estimated coefficients for OLS, fixed effects (LSDV) and random effects

OLS Fixed Effects (LSDV) Random EffectsExplanatory Variables

LNDIST -1.93*** (0.06) 7.47*** (1.85) -1.99***(0.14)LNPOP 0.80*** (0.18) 1.45*** (0.35) 0.80*** (0.04)BORDER 1.48*** (0.18) 18.63***(3.67) 1.53*** (0.54)OIC 0.03 (0.08) 4.11*** (0.82) 0.10 (0.17)LNGDPOCAP 0.93*** (0.06) 0.89*** (0.22) 0.94*** (0.05)LNEXC -0.03 (0.46) 5.61 (5.91) -0.16 (0.30)LNROOM 0.51 (0.55) -3.52 (13.89) 0.62* (0.35)LNALLOC 0.16 (0.23) 0.70 (1.77) 0.30** (0.15)Constant -4.45* (2.35) -70.09 (0.66) -8.04*** (2.08)Descriptive statistics and testsAdjusted R2 0.8120 0.9334 0.8111P-value 0.0000 0.0000 0.0000

Note: *** , ** and * indicate that coefficients are significant at 1, 5 and 10 percent levels respectively.

3 OIC is the largest intergovernmental organization in the world after the United Nations (UN). This organization has 57 member countries with the objective to strengthen solidarity and cooperation among them.

076.indd 47 3/27/18 3:07 PM



with the objective to attract more Muslim tourists to Malaysia. However, the government should not only focus on these Muslim-friendly programs to avoid crowding out arrivals from non-Muslim countries.

ConclusionTourism industry can be classified as one of the services sector and today, this industry has become an important industry and has grown to be among the largest industry after automobiles and oil in terms of revenue (Mohd Hanafiah & Harun, 2010). Therefore, this study was conducted to identify the factors and estimate their importance in explaining tourist arrivals to Malaysia in order to understand the decision-making behavior of tourists.

This study employed the bilateral tourism flows gravity model. The bilateral tourism flows gravity model was estimated using Least Squares Dummy Variable (LSDV). The model includes population, distance, income level of origin country and exchange rate to control for international shocks. This study also includes supply factor which is number of hotel room availability. Supporting variables like commonality in colonizer between Malaysia and sources of origin of the tourists was also examined. Furthermore, membership in the OIC is also used to determine the Muslim country.

The empirical analysis shows that population of country origin, country that share common border with Malaysia and OIC country are major factors that increase international tourist arrivals. Furthermore, the result also explains that the larger the market size of tourist’s country of origin, the more tourists tend to visit the destination country. From the result, it shows that distance is positive and statistically significant, however country specific effect distance is negative. The fixed effect shows the importance of specific country effect. Therefore, more data is needed to determine the effect of distance. Furthermore, country with zero arrival are not included in the regression which is one of the problem with tourist arrival data.

AcknowledgementsThis study is financially supported by Ministry of Education Malaysia, Niche Research Grant Scheme #NRGS13-003-0003 “Socio Economic Impact of Shari’ah Compliant Hospitality and Services on Malaysia and Muslim Communities”.

ReferencesAlgieri, B. (2006). An Econometric Estimation

of the Demand for Tourism: The Case of Russia. Tourism Economics, 12(1): 5-20.

Anderson, J. E., & Van Wincoop, E. (2003). Gravity with Gravitas: A Solution to the Border Puzzle. American Economic Review, 93: 170-192.

Bos, J. W. B. ve M. van de Laar. (2004). Explaining Foreign Direct Investment in Central and Eastern Europe: An Extended Gravity Approach, DNB Working Papers, 008, Netherlands Central Bank.

Deluna Jr, R., & Jeon, N. (2014). Determinants of International Tourism Demand for the Philippines: An Augmented Gravity Model Approach.

Dritsakis, N. (2004). Cointegration Analysis of German and British Tourism Demand for Greece. Tourism Management, 25: 111-119.

Egger, P. (2000). A Note on the Proper Econometric Specification of the Gravity Equation. Economics Letters, 66(1): 25-31.

Eichengreen, B., & Tong, H. (2007). Is China’s FDI Coming at the Expense of other Countries? Journal of the Japanese and International Economies, 21: 153-172.

Eilat, Y., & Einav, L. (2004). Determinants of International Tourism: A Three-dimensional Panel Data Analysis. Applied Economics, 36(12): 1315-1327.

Fourie, J., Rossello, J., & Santana-Gallego, M. (2015). Religion, Religious Diversity and Tourism. Kyklos 68(1): 51-64

076.indd 48 3/27/18 3:07 PM



Ghani, G. M. (2016). Tourist Arrivals to Malaysia from Muslim Countries. Tourism Management Perspectives, 20: 1-9.

Gil-Pareja, S., Llorca, R., & Martínez, J. A. (2007). The Effect of EMU on Tourism. Review of International Economics, 15: 302-312.

Hanafiah, M. H. M., & Harun, M. F. M. (2010). Tourism Demand in Malaysia: A Crosssectional Pool Time-series Analysis. International Journal of Trade, Economics and Finance, 1(1): 80-83.

Head, K., & Ries, J. (2008). FDI as an Outcome of the Market for Corporate Control: Theory and Evidence. Journal of International Economics, 74: 2–20.

Karemera, D., Oguledo, V. I., & Davis, B. (2000). A Gravity Model Analysis of International Migration to North America. Applied Economics, 32: 1745-1755.

Khadaroo, J., & Seetanah, B. (2008). The Role of Transport Infrastructure in International Tourism Development: A Gravity Model Approach. Tourism Management, 29: 831-840.

Kosnan, S. S. A., Ismail, N., Kaliappan, S. (2013). Determinants of International Tourism in Malaysia: Evidence from Gravity Model. Jurnal Ekonomi Malaysia, 47(1): 131-138.

Kusni, Aswad., Kadir, Norsiah., & Nayan, Sabri. (2013). Determinants of Tourism Demand in Malaysia by Tourists from ASEAN Countries: A Panel Data Econometric Analysis. International Academic Conference (IAC) 2013.

Lee, C. K., Var, T., & Blaine, T. W. (1996). Determinants of Inbound Tourist Expenditures. Annals of Tourism Research, 23(3): 527-542.

Li, G. (2004). Tourism Forecasting-An Almost Ideal Demand System Approach, Unpublished Ph.D Thesis, University of Surrey.

Malaysia Tourism Promotion Board. (2016). Tourism Malaysia. Retrieved from http://www.tourism.gov.my/media/view/malaysia-registers-25-7-million-tourists-in-2015 Accessed on 3rd August, 2016.

Massidda, C., & Etzo, I. (2012). The Determinants of Italian Domestic Tourism: A Panel Data Analysis. Tourism Management, 33(3): 603-610.

McCallum, J. (1995). National Borders Matter. Canada-U.S. Regional Trade Patterns. American Economic Review, 853: 615-623.

Mohd Salleh, N. H., Othman, R., Noor, A. H. S. M., & Hasim, M. S. (2010). Malaysian Tourism Demand from the Middle East Market: A Preliminary Analysis. International Journal of West Asian Studies, 2(1): 37-52.

Mohd Salleh, N. H., Siong-Hook, L., Ramachandran, S., Shuib, A., & Noor, Z. M. (2008). Asian Tourism Demand for Malaysia: A Bound Test Approach. Contemporary Management Research, 4(4).

Mohd Salleh, N. H., Cheah, C. Y., & Othman, R. (2011). Estimating Tourism Demand from China Market to Malaysia. Paper presented at the Business Innovation and Technology Management (APBITM), 2011 IEEE International Summer Conference of Asia Pacific.

Mohd Salleh, Norlida Hanim, Ahmad, S., Law, S. H., Zaleha, M. N., & Sridar, R. (2007). ARDL Model of Tourism Demand for Malaysia by Major ASEAN Countries. IIUM Journal of Economics and Management, 15(1).

Mohebi, M., Rahim, K. A. (2010). Revenue Determinants in Tourism Market. American Journal Applied Sciences 7: 1593-1598

Moorthy, R. (2014). An Empirical Analysis of Demand Factors For Malaysian Tourism Sector Using Stochastic Methods. Review of Integrative Business and Economics Research, 3(2): 255.

076.indd 49 3/27/18 3:07 PM



Morley, C., Rossello, J., Santana-Gallego, M. (2014). Gravity Modes for Tourism Demand: Theory and Use. Annals of Tourism Research 48: 1-10.

Munoz , T., & Amaral, T. (2000). An Econometric Model for International Tourism Flows to Spain. Applied Economics Letters 7: 525-529.

Qiu, H., & Zhang, J. (1995). Determinants of Tourist Arrivals and Expenditures in Canada. Journal of Travel Research, 34(2): 43-49.

Rodrigue, J. P. (2004). The Geography of Transport Heights. World Travel and Tourism Council. London: Routledge

Rose, A. K. (2000). One Money, One Market: Estimating the Effect of Common Currencies on Trade. Economic Policy, 20: 9-45.

Saray, M., & Karagöz, K. (2010). Determinants of Tourist Inflows in Turkey: Evidence from Panel Gravity Model. ZKU Journal of Social Sciences, 6(11): 33-46.

Seddighi, H. R., & Shearing, D. F. (1997). The Demand for Tourism in North East England with Special Reference to Northumbria: An Empirical Analysis. Tourism Management, 18(8): 499-511.

Song, H., Li, G., Witt, S. F., & Fei, B. (2010). Tourism Demand Modelling and Forecasting: How Should Demand be measured? Tourism Economics, 16(1): 63-81.

Tang, C. F. (2014). Assessing the Tourism-led Growth Hypothesis and the demand for tourism in Malaysia/Tang Chor Foon (Doctoral dissertation, University of Malaya).

Tinbergen, J. (1962). Shaping the World Economy: Suggestions for an International

Economic Policy. New York: The Twentieth Century Fund.

Var, T., Mohammad, G., & Icoz, O. (1990). Factors affecting international tourism demand for Turkey. Annals of Tourism Research, 17(4): 606-610.

Wang, Z. (2010). Factors that Influence the Growth of Chinese Domestic Tourism Arrivals (1985–2007) – An Empirical Research Based on the VAR model. Asia Pacific Journal of Tourism Research, 15(4): 449-459.

Witt, S. F., & Martin, C. A. (1987). Econometric Models for Forecasting International Tourism Demand. Journal of Travel Research, 25(3): 23-30.

World Tourism Organization, UNWTO. (2016). Press Release. Retrieved from http://media.unwto.org/press-release/2016-01-18/international-tourist-arrivals-4-reach-record-12-billion-2015, accessed on: 3rd August 2016

World Travel & Tourism Council. (2014). Travel & Tourism: Economic Impact 2014 world. Retrieved from http://www.wttc.org/~/media/files/reports/economic%20impact%20research/country%20reports/malaysia2014.ashx, accessed on: 17rd September 2015

Zaman, K., Khan, M. M. K., & Ahmad, M. (2011). Exploring the Relationship between Tourism Development Indicators and Carbon Emissions: A Case Study of Pakistan. World Applied Sciences Journal, 15(5): 690-701.

Zhang, J., & Jensen, C. (2005). Comparative Advantage in Tourism: A Supply-side Analysis of Tourism Flows. In: Proceeding of the Congress of the European Regional Science Association.

076.indd 50 3/27/18 3:07 PM

malaysia’s tourism demand: a gravity model...

Documents