1
Factors Influencing Tourism Demand to Indonesia
MuhamadYunanto
,
Henny Medyawati
FakultasEkonomi, UniversitasGunadarma, Depok
ABSTRACT
Indonesia is one country in the World that has many potential places that have to be visited by foreign tourists. Tourism has become one of the largest industries in the world. For Indonesia, tourism has an important role in increasing foreign exchange. The main objective of this study was to analyze the factors that influence the demand for Indonesian tourism. This study uses secondary data from the period 2010 to 2016. The number of sample applied are 16 countries selected by ranking the number of tourists most. The data used is the number of foreign tourists coming to Indonesia, the national income per capita of 16 countries, the exchange rate of 16 foreign currencies against the rupiah and tourism price. The method of estimation conducted using panel data regression. The empirical results applied a fixed effect model showed that the national income per capita, the exchange rate and the price of tourism affects the number of tourists coming to Indonesia.
Keywords:tourism, panel data, tourism price
INTRODUCTION
The Indonesian economy in 2016 remains resilientin the midst of global economic conditions that are still not strong and full of uncertainty. This development is influenced by dominant domestic demand structure and supported by adequate policy response.Indonesia's economic growth in 2016 increased from 4.9% in 2015 to 5.0% (Bank Indonesia, 2016).Indonesia's economic growth since 2010 is as follow; the Indonesian economy grew 6.1% in 2010, higher than 4.6% year-on-year growth. Indonesia's economy in 2010 continued to improve, supported by solid domestic demand and conducive external conditions. Global economic recovery that gradually began to occur since the first half of 2009 still continues in 2010, supported by high economic growth in emerging market countries (Bank Indonesia, 2010). Indonesia's economic growth in 2013 is in a slowing trend influenced by unfavorable global conditions and unfavorable domestic economic structures. The dynamics of the global economy also affected the economic performance in the form of economic growth trends that have slowed since the first quarter, so for the whole year it was 5.8%, slowing from the 2012 growth of 6.2%. Meanwhile, hotel and restaurant sub-sectors grew significantly in response to the increasing number of tourist arrivals and increased election activity in the second half of 2013 (Bank Indonesia, 2013). The development of tourism has an important role in encouraging economic activity, improving the image of Indonesia, improving the welfare of the community, and providing expansion of employment opportunities. Tourism has an important role in increasing the country's foreign exchange by seeking an increase in the number of foreign tourists (tourists) and an increase in the average expenditure of foreign tourists in Indonesia (KementrianPariwisatadanEkonomiKreatif, 2012).
The development of the tourism sector in Indonesia is shown, among others, from the number of tourist arrivals which tended to increase from 5,506 million in 2007 to 6,234 million in 2008, or grew by 13.2% and increased by 1.4% to 6,324 million in 2009 (Nizar, 2011). The number of foreign tourists visiting in 2012 was 8.04 million or grew by 5.16% when compared to the number of visits in 2011 of 7.64 million visits by foreign tourists (Ministry of Tourism and Creative Economy, 2012). In 2009, Indonesia was able to absorb about 0.72% of the world's tourist arrivals.Contribution of Indonesian tourism industry to Gross Domestic Product (GDP) in 2008, already equal to Rp. 153.25 trillion or 3.09% of the total GDP of Indonesia (Central Bureau of Statistics-BPS, 2014). The condition of tourism in 2014 on a macro basis, shows the contribution of tourism to the national GDP of 4.01%, foreign exchange generated reached US $ 11.17, and tourism labor as much as 10.32 million people. In the micro condition, the number of foreign touristsare as much as 9.44 million tourists, and domestic tourists (wisnus) as much as 251,20 million trips. For competitiveness, tourism in Indonesia according to the WEF (World Economic Forum) position ranked at 70 from the whole world (Ministry of Tourism, 2014). The role of tourism to Indonesia's GDP growth and its role as one of the contributors to export earnings, make tourism an important factor for the Indonesian economy.The purpose of this study is to analyze the factors that affect the demand for tourism in Indonesia as a tourist destination by foreign tourists. This research develops previous research from Yunanto and HennyMedyawati (2016) that is the addition of the number of years of research which previously 5 years to 7 years. Thereason is that this research could be more precise and accurate research in analyzing tourism demand in Indonesia. All variables used in this study adopted the research of Abdurrahim (2014), Yunanto and Henny Medyawati (2016) who also adopted the model of Munoz and Amaral (2000) that is the variable number of foreign tourists who come to Indonesia, the exchange rate of foreign currency against rupiah, per capita national income and tourism prices.The contribution of this research is the empirical findings in the development of models as well as donations or considerations for the government in formulating policies related to tourism.
LITERATURE REVIEW
The following are descriptions of previous research results related to the tourism sector and macroeconomic indicators such as inflation and per capita national income. Harun, Mohd.HafizdMohd., HanifahMohd., and FauziMohd. (2010), examine the demand for tourism in Malaysia based on key economic factors such as income, prices, exchange rates, consumer price index, population and economic crisis using modified Gravity models. Movements, patterns and changes from international tourist arrivals are also analyzed.The sample countries in this study were Australia, Hong Kong, Indonesia, Great Britain, Thailand, Taiwan and China. Estimation method using panel data regression. The results show that although Malaysia is experiencing economic crisis, Malaysia can still rely on tourism industry means to maintain the economy. Moorty (2014) also examines the demand for tourism in Malaysia by using gravity model and panel data in 2014 simultaneously with the Visit Malaysia Year 2014 program. The results show that the population, the distance between countries, the country directly adjacent to Malaysia and the country with a cognate language affecting Malaysia's foreign exchange.
Nizar (2011) examines the influence of tourism on economic growth. This research then continued by analyzing the number of tourist and foreign exchange to the Rupiah exchange rate in Indonesia in 2014. Both research done by using quantitative approach, that is autoregressive vector model (VAR) and using monthly time series data. The results show that tourism affects economic growth (Nizar, 2011) and tourism growth (tourism and tourist numbers) and exchange rates have reciprocal causal relationships.This is as a result of the increase of tourism foreign exchange which increases (appreciation) of Rupiah exchange rate for 3 months, while the increase of tourist number will increase (appreciation) of Rupiah exchange rate for 8 months. The results also show that the appreciation (depreciation) of the Rupiah will encourage the increase (decrease) of foreign exchange tourism and the number of tourists in different time and there is a positive relationship and mutual influence between the number of tourists and foreign exchange tourism (Nizar, 2014). Ibrahim (2011), analyzed the factors influencing the influx of tourist arrivals to Egypt. The results of the research with panel data regression show that GDP per capita, consumer price index (CPI), trade volume of Egyptian countries with 8 sample countries, and relative living cost of tourists and CPI of Tunisia as competitor country, affect the number of tourist arrivals to Egypt. Abdurrahim (2014) analyzed the role of national income, tourism prices, and exchange rates against Chinese tourist visits to ASEAN countries, during 2007-20014. The results obtained using the fixed effect model show that China's national income significantly and positively influence the visit of Chinese tourists to ASEAN.Putri (2014) analyzed the factors influencing tourism demand to Indonesia by using 21 sample countries. The results show that tourism prices, national income per capita, exchange rate, and distance between countries affect the number of tourists. The difference of this research with previous research is that this research does not use dummy variable as control variable, considering the condition of economic growth rate in 2010 that is 6.1% is at relatively high level compared to previous years.
RESEARCH METHOD
This study uses secondary data from 2010 to 2016 period. Data are sourced from BPS, Bank Indonesia, World Bank, International Monetary Fund (IMF) and Bank of Canada. The complete operationalization of variables can be seen in Table 1. below.
Table 1. Research Variables
No. / Variable / Description1. / jlh_wis / Number of foreign tourists
2. / GNI / Gross National Income(GNI) per-capita
3. / ex_rate / The rupiah exchange rate of 16 tourists countries of origin
4. / Hrg_par / Tourism prices
Data on the number of foreign tourist arrivals to Indonesia 2010-2016 period is obtained from the Ministry of Tourism website that is the data of monthly foreign tourist visits on Tourist Statistics Abroad. The selected countries to be sampled are 16 countries based on the ranking of the largest number of visits, namely Singapore, Malaysia, Australia, Japan, China, South Korea, USA, UK, Philippines, Germany, Netherlands, France, Hongkong, Thailand, India and Russia.National Income Data is measured with a GNIper capita in one million USD earned from the World Bank page. Indonesia's tourism price data is derived from the comparison of the Consumer Price Index (CPI) of other countries to the Indonesian CPI. The CPI of the 16 sample countries was obtained from the IMF page.The model used in this study adopts with modification of Munoz and Amaral (2000) research model, which is to analyze the demand for international tourism to Spain. Models of Garin and Munoz which are also adopted by Princess (2014) are as follows:
Ltourit = α + β1LGNP i,t + β2LEX i,t + β3LPR i,t + β4 D91t + µit
Notes:
- Ltour it is the log of the number of night spend in Spain from the tourists from Country i within the period t
- LGNP i, t is the log of State GNP i within t period
- LER i, t is the log of the Spanish currency exchange rate against the country currency i within the period t
- LPR I, t is the log of the tourism price index in Spain divided by the State CPI i within the period t
- D91 is a dummy variable that aims to represent the Gulf War event in 1991
- μit is an error variable
The model used in this study, adopted models from Munoz and Amaral (2000) but did not use dummy variables. The consideration of the unused dummy variable in this study is considerring to the Economic Report of 2010, that Indonesia's economic growth reaches 6.1% and the global economy in a conducive condition (Bank Indonesia 2010). This is also the reason for choosing 2010 as the year of study. The model is:
Log_wis = α + β1Log_GNIi,t + β2Log_EXi,t + β3Log_PR i,t + µit
The estimation method used in this study is panel data regression. Panel data is best for detecting and measuring impacts that simply can not be seen in pure cross-section data or pure time-series (Gujarati and Dawn C. Porter, 2012). The first step of data processing in this study is to test the classical assumption, which aims to ensure that the obtained model really meets the basic assumption in the regression analysis which includes assumption of normality, no autocorrelation, no multicollinearity and no heteroscedasticity.The next test is the Mackinnon, White and Davidson (MWD) method to determine whether the model is linear or log-linear. The next step is to make estimations using Common Effects, Fixed Effects and Random Effects. Model selection between Common Effect and Fixed Effect is done through Chow test or likelihood ratio test. The next step of model selection between Fixed Effect and Random Effect is done through Hausmann test. Then proceed with LM test for model selection between Random Effect and Common Effect.
RESULT AND DISCUSSION
At the beginning of the discussion described the condition of variables used in this study during the study period that is from 2010 to 2016. Based on the Ministry of Tourism report (2014), GDP generated from national tourism has increased significantly every year. In 2010 tourism generated GDP of 261.06 trillion rupiah and increased in 2011 to 296.97 trillion rupiah, in 2012 amounted to 326.24 trillion rupiah, and in 2013 the value of GDP generated reached 365.02 trillion rupiah.Furthermore, in 2014, the GDP generated from the tourism sector is expected to reach 391.49 trillion rupiah. Overall the total visit of foreign tourists for 16 countries in 2016 reached 8,299,056 visitors, in 2015 reached 7,705,916 visitorsand in 2014 reached 8,170,808 visitors. Here are the numbers of foreign tourists who come to Indonesia from 2010 to 2016 as shown in Figure 1.
Figure 1. Number of foreign tourist to Indonesia
The largest number of foreign tourists visit to Indonesia from 2010 to 2016 is: Singapore as many as 10,862,915 (21.06%), then Malaysia 9,304,604 (18.04%), Australia 7,040,187 (13.65%), China as many as 6,058,803 (11.70%) and Japan as many as 3,341,036 (6.43%).The next description is the condition of GNI per capita from the 16 countries during 2010 to 2016, as shown in Figure 2.
Figure 2. GNI per capita from 16 countries
GNI per capita in East Asia and the Pacific in 2016 shows an increasing trend even though some countries, including Indonesia, are slowing down. In fact, when compared to other Asean member countries, Indonesia's GNI per capita growth is recorded as the only country that incised GNI contraction based on data compiled by the World Bank. Indonesian GNI per capita is down 2.93% from US $ 3,760 in 2013 to US $ 3,650 over the past year. In fact, in 2012-2013 GNI Indonesia is still growing 4.44% from US $ 3,600 to US $ 3,760. With the GNI range, Indonesia is still included in the category of middle-income countries with 50 other countries whose data are recorded by the World Bank. The performance of the GNI is contradictive to the trend of GNI growth in the region,which is with a positive tone. There are other factors affecting GNI in addition to growth, i.e. the exchange rate is therefore measured in US dollars (Bisnis.com, 2015). The next description is the currency exchange rate variable of 16 countries with rupiah currency, as shown in Figure 3. below.
Figure 3. The Rupiah Exchange Rate in the period 2010-2016
The rupiah exchange rate especially against to the US dollar since 2010 continues to depreciate, which is from the range of 7,500 in 2010 to almost 13,400 in 2016. The weakening rupiah exchange rate not only occurs in USD currency, but also with other hard currency such as Japanese Yen, Singapore Dollar, and Euro. This condition also occurs to the eyes of other ASEAN countries such as Malaysian Ringgit, and Thai Baht.Based on the Economic Report of 2014 (Bank Indonesia, 2014), point-to-point, the rupiah depreciated by 1.7% to reach Rp11,876 per US dollar lower than the 20.8% depreciation in 2013. Meanwhile, on average, the rupiah weakened by 12.0% accompanied by steady declining volatility. The next discussion about the CPI growth of 16 countries, shown in Figure 4.
Figure 4. The CPI of 16 Countries in the period 2010-2016
The CPI condition of 16 sample countries shows a relatively stable condition during the study period. Japan in 2010-2012 even experienced minus inflation. This condition is different from Indonesia, where the government has issued several policies to control inflation. The government issued several price adjustment policies, particularly in the energy sector, to reduce the burden of subsidies to be diverted to development financing and more productive sectors. Gradually, the Government adjusted the price of LPG 12 kg and electricity tariff (TTL) for certain customer groups. The policy of adjusting subsidized fuel prices starting November 18, 2014 and food price volatility then caused the overall inflation of 2014 to be above the set target. The inflation rate of 2014 reached 8.36% (yoy), above the inflation target of 2014 (4.5 ± 1%) (Bank Indonesia, 2014).Indonesia's improved economy in 2016, also supported by the controlled of inflation. Inflation in 2016 was recorded quite low at 3.02%, thus continuing the achievement of 2015, which is within the 4.0 ± 1% range target. Achievements are influenced by low commodity prices, controlled exchange rates, well managed aggregate demand, and lower inflation expectations. These factors contribute to inflationlow core i.e. 3.07%.
The next step in this research is the data test. Before making an estimate, the important step that must be done is testing the classical assumption as a requirement for panel data regression. Based on the normality test that has been done, it can be concluded that the data of the number of tourists, national income per capita, the rupiah exchange rate of 16 countries of origin of tourists and the price of tourism meet the assumption of normality. The next step is multicollinearity test done to know whether or not there is deviation of classical assumption of multicolinearity. The test results show that the Tolerance value of the three variables is more than 0.10 and the VIF value is less than 10 thus there is no multicollinearity problem in the regression model.Heterocedasticity tests were used to test whether in the regression model there was a residual variant inequality for all observations. Heterokedastisity test can be seen mellui graph of plot between variables with residual. If there are certain patterns, such as the points that make up a particular pattern on the scatterplot (wavy, widened, then narrowed), then there has been identified heteroskedasitas. If there is no clear pattern, as well as spreading points above and below the zero on the Y axis, no heteroscedasticity occurs. The result of autocorrelation test of DW value equal to 1,457. The DW value is between -2 and +2. This shows that there is no autocorrelation.
Before the panel data regression is done, then the data should be tested with MWD method. The first step by doing linear regression to get residual value which further residual value is stored (Res1 = resid). The next process is to form a new variable F1 = Y-Res1. The next step is regression of linear log model, also to get residual value (Res2 = resid) and formed new variable F2-logY - Res2.Variabelbaruberikutnya yang perludibuatadalah Z1=logF1-F2 dan Z2=antilog(F2 – F1). Further regression is conducted by entering the variable Z1 on the regression of linear equations and Z2 on linear log equation regression. Regression results showed that both linear and logical models showed equally good results, with a high coefficient of determination R2 for each equation of 0.990 and 0.997. This study further uses the loglinier model, referring to the model adopted from Munoz and Amaral (2000).