The Effect of Number of Visitors, Tourist Destinations, Hotel Room Tax and Accommodations on Original Local Government Revenue: Case Study West Sumatra Province, Indonesia

This research aims to discover 1) The effect of number of domestic visitors on Original Local Government Revenue (OLGR) 2) The effect of number of foreign visitors on OLGR 3) The effect of number of tourist destinations on OLGR 4) The effect of restaurant tax on OLGR 5) The effect of hotel room tax on OLGR 6) Number of accommodations as a moderating variable for relationship hotel room tax and OLGR. The study population consisted of 12 regencies and 7 municipalities. The sampling technique uses purposive sampling. The selected sample is considered the most appropriate to represent tourism according to Tourism Office of West Sumatra Province. The selected sample is 3 municipalities and 2 regencies. Data source obtain from Central Bureau of Statistics (BPS) West Sumatra Province. Data analysis consisted of statistical descriptive analysis, model estimation test, classical assumption test, coefficient of determination test, F-test and t-test. The results show 1) The number of domestic visitors has a positive and significant effect on OLGR 2) The number of foreign visitors has a positive and significant effect on OLGR 3) The number of tourist destinations has a positive and significant effect on OLGR 4) Restaurant tax has a positive and significant effect on OLGR 5) Hotel room tax has a positive and significant effect on OLGR 6) Number of accommodations show evidence as a moderating variable for relationship hotel room tax and OLGR.


Introduction
Regional autonomy policy can have a positive effect on the regions in terms of regional sovereignty to regulate their domains. When compared to a centralized system, this policy is considered to be superior because the regions are the main actors in development and are not side players. Since the implementation of regional autonomy, this policy had a significant effect on the regions to maximize the existing potential due to the implementation of decentralization. Decentralization policies also have an impact on a country's economic growth and reduce poverty (Feltenstein & Iwata, 2005); (Condro et al., 2019) Local governments need to encourage community-based economic sectors or local revenue optimization. The central government should make the municipalities and regencies better able to concentrate on empowerment of local economic power so that the direct impact of economic growth, in addition to the increase in local revenue is also felt directly by the community. One way to grow the regional economy is to improve tourism governance (Koster, 2008); (Archabald & Naughton-Treves, 2001); (Walpole & Goodwin, 2000); (Blom, 2000); (Spenceley et al., 2019). years has increased because the local government makes efforts to increase tourism. Developments can be seen from the increase in the number of accommodations that occur every year in the form of hotels and other accommodations.

Literature Review
Original Local Government Revenue (OLGR) is revenue that is withheld based on local regulations under the legislation to finance their activities. OLGR consists of 3 main aspects, including local taxes, retributions, and income of regional government corporate and management of separated regional government wealth. The relationship between the tourism industry and regional revenue is connected through the regional revenue channel and tax revenue sharing. The tourism sector can be a linkage for other product and service subsectors in increasing Gross Regional Domestic Revenue (GRDP) and local revenue itself.
Previous research by (Parida et al., 2017); (Sheng, 2017) stated that domestic visitors and foreign visitors can improve the local economy through revenue receipts from tourism. Domestic and foreign visitors are usually attracted to tourist objects, so the increase in the number of admission tickets will increase OLGR. Previous studies by (Nicely & Palakurthi, 2012); (Wall & Zhao, 2017) stated local revenue will increase due to visitor average daily spend. These expenses include consumption, accommodation, transportation, and telecommunication costs The higher the number of tourists, the higher the potential for tourists to extend their stay.
Previous studies by (Botti et al., 2018); (Gonzá lez et al., 2019) and (Santos et al., 2020) showed tourist destination increase local revenue. Regions that have high tourist destinations tend to have high local income potential. Every tourist destinations has something that can attract tourists. Objects can offer shows, natural beauty, shopping for souvenirs or culinary tours. In terms of culinary, for example, tourists tend to adapt to local food change their appetite (Santos et al., 2020). Local cultural performances are attractive to foreign tourists and are able to improve the regional economy (Rahmanita, 2019). Regions that have high tourist destinations have the potential to get higher local income.
Previous studies by (Sheng, 2017); (Afonso, 2015); (Bartle et al., 2003); (Gonzá lez et al., 2019); (Santos et al., 2020); (Bartle et al., 2003) showed hotel tax increase own-source revenue. In local tax revenue, hotel tax plays an important role because it is the highest contributor in several municipalities and regencies in West Sumatra. Although there are still many phenomena due to the constraints of the tax collection system that is applied to hotel taxes, considering the self-assessment system that requires honesty of the taxpayer itself. As a result, several regencies have not succeeded in exceeding the tax targets set. However, this indicates that the West Sumatra tourism sector has a positive influence on the development of West Sumatra in the future.

Methods
The research approach used is the causality approach. Causal research is research that has the main goal of proving a causal relationship or relationship affecting the variables studied. The research data used are secondary data related to the studied variables obtained from the Central Statistics Agency (BPS) and the Department of tourism during 2014-2019. The objects of this research are regencies and municipalities located in West Sumatra. The total population in this study was 19, which consisted of 12 regencies and 7 municipalities. The sampling technique is using purposive sampling, where the selected sample is representing West Sumatra tourism according to the Department of Tourism. The samples are Padang city, Bukittinggi city, Payakumbuh city, South Pesisir regency, and Mentawai Islands regency. Sample selection can be seen in Table 1 below.
Data analysis methods consist of descriptive analysis, determination of estimation models, classic assumption test (normality test, heteroskedasticity test, multicollinearity test, and autocorrelation test). Moderated regression analysis by using EViews program. To test the hypotheses, the F-test and the t-test were performed.

Number of Domestic
Visitors (X 1 ) Hotel Room Tax (X 5 ) Moderated regression analysis equation model is Y = α+β 1 X 1 +β 2 X 2 +β 3 X 3+ β 4 X 4 + β 5 X 5 + β 6 X 6 + β 7 X 5 X 6 +ε  Table 3, prob. value of Cross-section Chi-Square is 0.0129 < 0.005, which means that fixed effect is better than common effect to estimate the model. The next step is Hausman test. The result is shown in Table  4 above.
Based on Hausman test in Table 4  In classical assumption test, normality test use Jarque-Berra test. The result are shown in Figure 2 above. Based on normality test result, Jarque-Berra prob. is 0.474490 > 0.05, which means the data distribution is normal. The second classic assumption test, heteroskedasticity test.
The second classical assumption test, heteroskedasticity test use Glejser test. The result is shown in Table 5 below.
The results of the significance of each independent variable > 0.05. From this result, it can be concluded that there are no symptoms of heteroscedasticity in the data.   Data processed by author Moderated regression analysis was performed to determine whether the relationship between two variables depends on (is moderated by) the value of a third variable. The coefficient of determination test, F-test, and t-test using fixed effect estimation are shown in Table 8. Data processed by author The coefficient of determination test is presented in table 8 above. The value of Adjusted R Squared is 0.980497 or 98.0497%. That means the contribution of independent variables and moderating variables (Domestic Visitor (X 1 ), Foreign Visitor (X 2 ), Tourist Destination (X 3 ), Restaurant Tax (X 4 ), Hotel Room Tax (X 5 ) and Accommodations (X 6 ) on independent variable OLGR (Y) is 98.0497%. The rest is influenced by other variables outside this research.
To test the effect of the independent variable on the dependent variable partially, t-test was performed. The result is shown in Table 8 above. For H1 test, Domestic Visitor (X 1 ) prob. is 0.0023 < 0.05, t-statistic > t-table (4.505251 > 2.0687). That means Domestic Visitor (X 1 ) has a positive and significant effect on OLGR. H1 is accepted. This result is in line with previous studies by (Nicely & Palakurthi, 2012;Parida et al., 2017;Sheng, 2017;Wall & Zhao, 2017). An increase in the number of domestic visitors will increase local income. Tourists will purchase tickets to enter tourism objects, where the increase in ticket sales will increase OLGR. In this case, ticket sales for tourist attractions are a component of retribution that can increase OLGR Other purchases related to consumption during the trip will increase OLGR through restaurant tax revenue. Transportation costs paid by visitors will also increase OLGR because visitors need transportation from one tourist attraction to another. The high number of visitors will increase sales in shopping center-based attractions.
For H 2 test, Foreign Visitor (X 2 ) prob. is 0.0962 < 0.10, t-statistic > t-table (2.164661 > 1.7139). That means Domestic Visitor (X 1 ) has a positive and significant effect on OLGR. H 2 is accepted. This research in line with previous studies by (Nicely & Palakurthi, 2012;Parida et al., 2017;Sheng, 2017;Wall & Zhao, 2017). The explanation about domestic visitors also applies to foreign visitors. Although the number of foreign visitors is not able to equal the number of domestic visitors. In this case, the foreign visitor variable should not be ignored in increasing OLGR. They tend to like new things to add to tourism experience. Their presence tends to increase restaurant visitors and hotel occupancy rates. Things that according to residents or domestic tourists are common, on the contrary for foreign tourists it is an experience that has high value. Even in the literature that has been described, foreign tourists can change consumption tastes and adjust to local tastes. This will affect OLGR through local restaurant tax revenues.
For H 3 test, Tourist Destination (X 3 ) prob. is 0.0956 < 0.10, t-statistic > t-  (Botti et al., 2018); (Gonzá lez et al., 2019); (Santos et al., 2020); (Rahmanita, 2019). Tourist destinations can be based on natural tourism, marine tourism, historical tourism, cultural tourism, and culinary tourism. The more the number of tourist destinations and the more types of tourism offered to visitors, the visitors can have which alternative tours they prefer. Different types of tourist destinations will result in differences in tourist expenditure (Pratamawaty et al., 2019). OLGR will increase as the number of tourist destinations increases. This is due to the increasing number of tourist destinations, the higher the potential for tourist expenditure. Visitors who prefer culinary, they will be willing to pay more for culinary tours. Visitors who like nature tourism, they will stay longer and visit one place to another.
For H 4 test, Restaurant Tax (X 4 ) prob. is 0.0007 < 0.05, t-statistic > t- Eating local food contributes significantly to its overall tourism experience. Food curiosity makes visitors willing to spend more money to try something new that will add to their experience. The higher the number of visitors who change their appetite to local food, the restaurant tax revenue will increase along with food sales. Restaurant tax revenue will potentially increase OLGR. Several previous studies have also shown that local food will be preferred by visitors rather than serving the original food of the visitor.
For H 6 . Hotel room tax*number of accommodations prob. is 0.0074 < 0.05, t-statistic > t-table (3.019810 > 1.7139). the number of accommodations shows evidence as a moderating variable for the relationship of hotel tax with original local government revenue. H 6 is accepted. This research in line with previous studies by (Litvin et al., 2006); (Bird, 1992); (Dogru et al., 2020). In this study, we have found evidence of the number of accommodations that have a significant effect on OLGR. The increasing number of accommodations will increase tax revenue and have a positive impact on the increase in OLGR. Furthermore, number of hotels increases employment in both the overall economy and the tourism, leisure, and hospitality industries. Medium hotels make the highest contribution to employment in the overall economy. Small hotels make the biggest contribution to employment in the overall tourism, leisure, and hospitality industries.

Conclusion
This research aims to discover 1) The effect of number of domestic visitors on Original Local Government Revenue (OLGR) 2) The effect of number of foreign visitors on OLGR 3) The effect of number of tourist destinations on OLGR 4) The effect of restaurant tax on OLGR 5) The effect of hotel room tax on OLGR 6) Number of accommodations as a moderating variable for relationship hotel room tax and OLGR. The results show 1) The number of domestic visitors has a positive and significant effect on OLGR 2) The number of foreign visitors has a positive and significant effect on OLGR 3) The number of tourist destinations has a positive and significant effect on OLGR 4) Restaurant tax has a positive and significant effect on OLGR 5) Hotel room tax has a positive and significant effect on OLGR 6) Number of accommodations show evidence as a moderating variable for relationship hotel room tax and OLGR.
Local government and stakeholders need to improve all aspects of this research variable to maximize OLGR. For further researchers, it is recommended to add research variables related to tourism development. Researchers can expand the object of research and increase the number of samples using more specific criteria.