Multiplier Effects of Tourism Sector in Yogyakarta: Input-Output Analysis

In the Special Region of Yogyakarta (DIY), the tourism sector contributes roughly to 34% of the total output, either directly or indirectly. Because of the enormous potential of tourism in boosting the economy, the local government was encouraged to adopt plans and programs so that the tourism sector's long-term viability can influence the development of other sectors. This study aims to describe the economic multiplier effects of tourism industry operations. An examination of the input-output table of 35 sectors in the Special Region of Yogyakarta in 2015 was carried out in this study. An examination of output, income, gross value added, and labor linkage and multipliers was made from an input-output analysis. The hotel and restaurant industry, as well as air transport, is considered as the key sector in the linkage study (forward and backward linkage). The sector with the biggest employment absorption, according to the multiplier analysis, is trade.


INTRODUCTION
The tourism sector has both direct and indirect contribution approximately 8-9% to the economy of Gross Domestic Product (GDP) of ASEAN countries, including Indonesia, and creates 1 in 11 jobs (Antara, 2014). United Nation World Tourism Organization (UNWTO) estimates that the prospect of tourism, especially in ASEAN, is expected to grow to 10.3% in 2030 (UNWTO, 2018). According to the World Tourism and Travel Council, tourism sector in global GDP comprises 10.4%. The number of employees in this sector reaches 313 million and constitutes 9.9% of the total jobs in the world (WTTC, 2018).
A different situation occurred after WHO (12 March 2020) declared COVID-19 as a pande-mic, which significantly affected the global economic, politics, and social systems. Health communication strategies and measures, such as social distancing, travel and mobility bans, community lockdowns, stay-at-home campaign, mandatory self-quarantine, and gathering restrictions, have put a stop to global travel, tourism, and recreation (Sigala, 2020).
The tourism sector eventually became of the sectors hit hardest by the COVID-19 pandemic due to the decrease in travel demand and supply, which resulted in the decrease of tourism prices. Hoque et al. (2020) suggested that the tourism sector currently has to face threats that have never existed before. Chanona et al. (2020) suggested that tourism demand has changed during the pandemic because the tourism sector puts people at risk of infection. It was reported that the number of international tourist arrivals has dropped due to Covid-19 with a significant reduction of -22% globally in quarter I of 2020. The Asia-Pacific and Europe areas have the most decline of -34. 6% and -19.1% reduction, respectively (webunwto.s3.eu, 2020). As a result, other sectors such as airlines, hospitalities, and MSMEs which are related to the tourism sectors have also impacted as well as those sectors show a minus growth during pandemic (Harchandani & Shome, 2021). Bakar & Rosbi (2020) stated that existing regulations affected the decline in consumers, loss of business income, and layoffs resulting in higher unemployment of a country. UNWTO predicted that global tourism would decrease by 20 to 30% until March 2020.
The tourism sector, which has been the pillar of the Special Region of Yogyakarta, has also been in shock with the social distancing regulation. Restrictions on outdoor activities and overseas travel decreased the income from transportation, accommodation, and tourism industry sectors. As a result, the economy of the Special Region of Yogyakarta contracted to minus 0.17% in the first quarter of 2020. The open unemployment rate as of February 2020 rose to 3.38% compared to February 2019, which was 2.86% (BPS, 2020). In decades, the Special Region of Yogyakarta has had the highest rate of poverty and inequality in Java Island, and with this shock, its condition could more exacerbate. Thus, the recovery of tourism sectors is prominent way to maintain the economy.
Tourism is a sector which has a multiplier effect, which is a situation where tourism development can influence the development of other supporting sectors, such as infrastructure, service, transportation, food and beverages, as well as the opening of wider employment opportunities. In addition, the tourism sector has a significant and sustainable relationship with food, agriculture, and small and medium industries, thereby creating synergies in regional development. Tourism has also been proven to help poor areas (Hjerpe, 2018;Ferrari et al., 2018). Tourism has direct and indirect effects on the economy. Tourists who spend their money in a hotel help create jobs directly and provide value-added in the hospitality industry. However, other industry such as agriculture/ farmers where the hotel obtains their food, the craft industry which supply souvenir, and the clothing industry which provide hotel equipment also gain the indirect effect from the hotel activity. Therefore, tourism has a large multiplier effect on other sectors and it is very important to the rest economy (Rusu, 2011). Kumara, Prastyo, & Rahayu (2021) estimated that during pandemic, Indonesia lost potential income from tourism by IDR 99.09 trillion and has impacted other sectors such as FnB sector by IDR -17.9 trillion dan trade by IDR -10.7 trillion.
In the context of tourism economy, a multiplier approach is usually adopted to examine the existence of economic impact, which is determined through input-output (IO) analysis (Archer & Fletcher, 1996;Wagner, 1997;Sinclair, 1998). The multiplier measures the impact of each increase in tourist spending (demand) on the economy, usually in terms of output, income, and employment. Tourism IO studies have been carried out in the national and regional levels, for instance, Siswahto & Muryani, 2020 (North Sulawesi); Hasanli & Baizakov, 2019 (Azerbaijan, Kazakhstan, Iran ); Tohmo, 2018 (Finland, regional input-output); Munjal, 2013 (India); Atan & Arslanturk, 2012 (Turkey), Kim & Chon, 2009 (Korea); Kweka et al., 2003 (Tanzania); and Tohamy & Swinscoe, 2000 (Egypt). However, the analysis of the multiplier effect of tourism in the Special Region of Yogyakarta comprehensively and the both direct and indirect effects on all sectors was still limited. Thus, this current study attempt to analyze the multiplier effect of tourism using input-output (IO) analysis to identify the impact of tourism development on the economy of the Special Region of Yogyakarta. First, a multiplier analysis was performed to examine the impact of the tourism sector on output, income, employment, and community income. It is expected that the development of tourism sector has direct positive effect to the other sectors's output as well as income and employment, vice versa. Second, an analysis of the forward-backward linkage between the tourism sector and other sectors was made. Third, an analysis of investment needs was carried out for tourism development in the future.

METHOD
The Special Region of Yogyakarta Input-Output Table of 2015 for 35 sectors was utilized as the analytical tool in this study. It consists of 17 economic sectors in total. The data of the tourism sector in the input-output table were taken from the 2015 Indonesia Standard Industrial Classification (ISIC) in the tourism sector, the Special Region of Yogyakarta Special Input-Output Survey (SKIO) of 2015, the National Socio-Economic Survey (SUSENAS), the Indonesia Tourist Survey (WISNUS), the Passenger Exit Survey (PES), and the Moving Position Data (MPD). After determining the output value of economic activities in the tourism sector, the output value aggregation was carried out to get the output value of the tourism sector.
The economic impact of tourism can be investigated by analyzing its impact on production growth, factors of production, and balance of payments (Mikic, 1988). To estimate the direct-indirect impacts and variations in tourist spending, a multiplier analysis is needed, which can be considered as a measure to estimate the total impact of tourism on an economy. Inputoutput (I-O) analysis is an appropriate technique to capture the total impact of tourism through estimating the multiplier (Mazumder et.al., 2011). Where, Xi is a supply come from domestic output; Mi denote supply come from imports for similar products; Xij stands for Intermediate request; Fi is a Final request; and Vi denote the Primary input. Tourism Sector in Yogyakarta: Input-Output Analysis Each row in table 1 was calculated using this following equation: Subsequently, equation (1) was written in the form of a general equation: , Where, Xij is a Provision of sector i used by sector j; Xi stands for Total output (domestic) of sector i; Fi symbolised for Final demand for sector i; and Mi is Imports in sector i. Then, it was derived into an algebraic equation: Instead, a general equation was used: , With Vj is Primary input (GVA) of sector The direct backward linkage of sector j can be seen from the sum of intermediate input coefficient value of sector j or the number of elements of matrix A in column j. The number of intermediate input coefficient of sector j shows the degree of direct backward linkage of sector j. The direct backward linkage index of sector j was obtained from the following formula: Where, DBLIj denote a Direct Backward Linkage Index of sector j and aij is intermediate input coefficient of sector j from the i-th sector.
In the forward linkage, which is a degree of sensitivity, this analysis was interpret-ed as follows: if the forward linkage of sector I is more than or equal to or less than one, the degree of forward linkage of sector I is more than or equal to or less than one average of another sector's linkage in general. The number of forward linkage index signifies the degree of linkage of a sector, which if it increases, it can encourage development in more dowFnstream sectors (as an input to other sectors or sales).
The direct forward linkage of sector i can be seen from the number of intermediate input coefficient value that is in line with sector i or the number of elements of matrix A in row i. The number of intermediate input coefficient of sector i indicates the degree of direct forward linkage of sector i. The direct forward linkage index of sector I was obtained from the following equation: Where DFLIi is Direct Forward Linkage Index of sector I; aij is intermediate input coefficient of sector j from the i-th sector.
Multiplier analysis attempts to see the impact of changes in final demand (exogenous variables) on sectoral output (endogenous variables) in the economy. The three multiplier analysis include output multiplier, household income multiplier, and employment multiplier. Output multiplier analysis was performed to identify the impact of changes in sectoral dem-and on the output of other economic sectors. When demand for one sector increases, it will have an impact on the increasing production in other sectors. Products from other sectors are needed as production inputs for the sector. In the IO analysis, output multiplier can be written as follows: (7) Where X is output matrix; F is exogenous variable; (I-A) -1 is multiplier matrix.
First, this study will estimate groos value added. According to Wikarya (2015), gross value added indicates the amount of remuneration from production factors, consisting of wages and salaries, business surplus, indirect taxes, depreciation, and subsidies, which can be formulated as follows: (8) Where, V 201j is expense or income for employees; V 202j is business surplus (profit and payment to the state); V 203j express the expense for depreciation of capital goods; V 204j stands for net indirect tax: net value added tax; and V 205j is government subsidies.
Gross value added coefficient of sector j was obtained from: After obtaining a number of Vj as many as the sectors in the IO table, they were placed into a diagonal matrix.
The sectoral gross value added was calculated using: Where, M is gross value added; is diagonal matrix of n x n size; while (1-A) is a multiplier; and F denote the stimulus.
Second, the multiplier estimated is household (community) income multiplier. Household income multiplier, or commonly referred to as income effect of the IO model, indicates changes in final demand which will have a direct or indirect impact on changes in income of the sector itself or other sectors depending on the income multiplier (Nazara, 2005). The income effect value was formulated as follows: Where, M is income effect matrix n x n size; (1-A d ) -1 denote the total output multipler matrix; and is income coefficient diagonal matrix n x n size.
If the income effect is the result of the diagonal matrix of income coefficient multiplied by output multiplier, the impact of final demand on changes in income becomes: Income multiplier of sector j was determined using the formula: yj = Where, yj is income multiplier of sector j; mij stands for element of income effect in matrix row i column j; and vj is income coefficient of sector j.
The number yj shows how much income changes for the economy as a whole if the income of employees in sector j changes by one unit of money.
Third, Employment multiplier is assessed. Employment multiplier or commonly referred to as employment effect, is the total effect of changes in employment in the economy due to an increase in final demand in a particular sector. Employment multiplier is similar to household income multuplier. The main difference is that employment multiplier is expressed in units of emplotment. Employment effect was formulated as follows: Where, E denote the employment impact matrix; and is an employee coefficient matrix consisting of the ratio of employee to the total input of each sector.
Employee coefficient matrix is a diagonal matrix whose components were obtained from: Where is the number of employe in sector j; and is total input of sector j.
Changes in the number of employees needed due to changes in domestic final de-Nuryadin & Purwiyanta, Multiplier Effects of Tourism Sector in Yogyakarta: Input-Output Analysis mand of each sector were formulated as follows: Employment multiplier of sector j was determined using: Where, Zj is employment multiplier of sector j; express the element of employment (E) impact matrix in row i column j; and is an employee coefficient j.
The number on Zj implies the number of addition (reduction) of employment opportunities for the economy as a whole if employment opportunities in sector j increaes (decreases) by one person.
To determine the impact of changes in final deman in the tourism business sector on the regional economy, especially regional economic development, an analysis was carried out by entering the shock value in the IO table so that the impact arising from changes in the output of tourism business sector was found to be interpreted. The data used to analyze the impact of tourism development included the data from local government budget (APBD) for tourism activity programs or tourism supporting activity programs in the Special Region of Yogyakarta, the data on private investment in the tourism sector, and the data on tourist spending. A sector is deemed crucial (leading) in the economy if it has a forward and backward linkage index of more than one (Didit & Devi, 2008;Siswahto & Muryani, 2020). The hotel and restaurant sector (28) and the air transportation sector (31) are at least two tourism-related sectors that are included in the key sectors. Despite not being a key sector, the land transportation sector (29) has the highest distribution power, indicating a 1.181 rise in the output of other sectors providing inputs (including the sector itself). The three industries that play a relatively substantial part in the formation of output are (32) other services, (28) hotels and restaurants, and (27) trade.

RESULTS AND DISCUSSION
The output multiplier in tourism-related sectors includes: (27) (27) trade sector signifies that an increase of 1 million in the final demand in that sector will raise the economic production (total of all sectors) by 1.1841 million, ceteris paribus. Investing or developing in sectors with a high output multiplier value will yield the best results (Appendix 2).
Sectors with high multipliers of gross value added (GVA) include: (27)  According to the calculation of the added value of tourism-related sectors, such as the hotel and restaurant sector (28) of 0.7541, an increase in the final demand for the hotel and restaurant sector (28) of IDR 1 million will increase the added value in the economy by IDR 0.7541 million, ceteris paribus. All variables included in the added value, such as wages and salary, corporate surplus, depreciation, and indirect taxes, will increase if the final demand rises in industries with high GVA multipliers. These results are in line with the previous literatures conducted in Indonesia and other countries (Kumara, Prastyo, & Rahayu, 2021). The results show that these sectors have a direct influence from tourism activities so that if a shock occurs, the sector will experience quite a severe impact.
In tourism-related sectors, the multiplier value of household income is as follows: (27) trade (0.2807), (28) hotels and restaurants (0.2497), (29) land transportation (0.2968), (30) water transportation (0); (31) air transportation (0.2363), and (35) other services (0.4757). Sector (35), other services, has the highest income multiplier, which is 0.4757, implying that an increase of IDR 1 million in the final demand in the sector will boost household income by IDR 0.4757 million, ceteris paribus. If the final demand for sectors with strong household income multipliers rises, household incomes in tourism-related sectors will rise as well.
The values of the labor multiplier in tourism-related sectors are as follows: (27) (27), trade, has the highest multiplier figure, which is 0.0400. It can be interpreted that an increase of IDR 1 million in the final demand for the sector will increase employment by 40 people, ceteris paribus. The injection (shock) for analysis of the economic impact of the tourism sector development in the Special Region of Yogyakarta were obtained from: (i) tourist spending in 2015-2020 and its trends; (ii) tourism sector investment (private/ community) in 2015-2020; and (iii) government spending (local government budget) for tourism investment and promotion in 2015-2020. With that assumption, the impact of tourism development in the Special Region of Yogyakarta in 2015-2020 on the regional economy could be calculated, both on output, value added, household income, and employment. The result of the calculation can be seen in the following figures 1.
In 2015, the tourism sector development in the Special Region of Yogyakarta could increase the GDP at current prices with IDR 16.563 billion or about 16.72 and absorb 298.5 6 employees or around 15.78%. With the same assumption (according to the trend), in 2016 -2019, the tourism sector development in the Special Region of Yogyakarta was estimated to increase the GDP at current prices about 20% each year and absorb employee by an average of 23% each year. A different condition occurred in 2020, where the tourism sector only had an impact of 6% and absorbed 8.5% employees. This was due to the COVID-19 pandemic as a consequence of the social restriction regulations which significantly affected the tourism sec-tor with the decrease in tourist spending. In reponse to the situation, the government of the Special Region of Yogyakarta refocused its budget on the tourism sector, which was declining by 70% from the previous year. For instance, the direct expenditure component of the Tourism Office of the Special Region of Yogyakarta was recorded to decrease from IDR 56.192 million in 2019 to IDR 17.721 million in 2020. This result resonates with previous findings which stated that there has been unexpected decrease in growth and employment about 2.1% to 6% in Greek and China (Mariolis et al., 2020); Shuifa et al., 2011). Furthermore, from the point of view of percentage of the impact of injection (shock) on GDP and the contribution of the tourism sector to the formation of GDP, the same pattern occurred in 2015-2019, which tended to increase, and contracted in 2020, where injection had an impact of only 6.09% and the contribution of the tourism sector was 21.31% or declining from 27.08% compared to 2019.

CONCLUSION
The role of the tourism sector in the economy of the Special Region of Yogyakarta has fluctuated, although not significantly, throughout the year, including in 2020. In 2016, the tourism sector contibuted to 26.94%, in 2017 27.21%, and in 2018 27.17%. However, there was a decrease in 2019 and 2020 by 27.08% and 24.09% respectively. The largest contribution came from accommodation and food and beverage at 9.98%.
The main tourism-related sector that has the largest multiplier value is the land transportation sector. The output multiplier value of land transportation sector is 1.545, the gross value-added multiplier 0.770, household income multiplier 0.296, and employment multiplier 0.009. Other sectors that have quite large multiplier values are air transportation, other services, hotels and restaurants, trade, and water transportation.
The impact of the COVID-19 pandemic was seen in 2020, where the percentage of the impact of the tourism sector was 6.09 on the economy and absorbed 8.5 labors. The contri-bution of the tourism sector was 21.31%, a decrease from 2019 which was 27.08.
The biggest contribution of the tourism sector on the regional economy of the Special Region of Yogyakarta is from the domestic tourist spending. Therefore, efforts are needed to increase the number of tourists (comfort) and unique and typical products from the Special Region of Yogyakarta to support its tourism.
Efforts to increase the contribution can also be made by facilitating business licensing and providing investment incentives to attract private investment in the tourism sector in the Special Region of Yogyakarta. Private investment should be prioritized over investment using local government budget (APBD).