The Determinant Factors of Automobile Production in Indonesia

Article Info ________________ History of Article Received October 2020 Accepted December 2020 Pusblished February 2021 ________________


INTRODUCTION
The largest transportation industry in the world is the automobile industry, with more than one billion units used globally. The automobile industry is one of the gates for innovation, technology transfer, infrastructure improvement, socio-culture shifting, and global market expansion which boost a nation's economy as the primary goal. The automobile industry utilizes high capital investment, absorbs significant employment, improve technology capability, and opens opportunities to develop related industries. Therefore, many developing countries eagerly promote their automobile industry through policies series (Jan & Hsiao, 2004). In addition, the automobile industry still positive growth globally for the past 15 years ( Figure 1). However, the typical characters of developing countries are lack of qualified human capital, technology, management skill, and global network information. These factors force developing countries to seek assistance from Multinational Companies (MNCs) to develop their automobile industry through MNCs' Foreign Direct Investment (FDI). FDI is an essential element in the automobile industry for developing countries and the main reason how and why the automobile industry exists in developing countries.
For many years, the Indonesian automobile industry is one of the industries which has received a large number of FDI and the Government of Indonesia (GoI) interventions. However, many studies point out that during the New Order era, the GoI failed to develop its automobile industry because of unconstructive regulations (Adnan, 2014;Okamoto & Sjöholm, 2000;Tarmidi, 2004).
As a result, as the fourth largest populated country in the world, Indonesia automobile production is the seventh rank among automobile producer developing countries (Table 1). Moreover, according to Gabungan Industri Kendaraan Bermotor Indonesia (Gaikindo), with more than 2 million units production capacity, Indonesian automobile industry only produces less than 60% of its production capacity indicating the industry does not optimally perform yet.
Since Indonesia has a large number of domestic market, low-cost capital resources, growing middle-class income society, automobile MNCs still has the interest to put their investment to Indonesia (Natsuda & Otsuka, 2014). However, FDI inflow is still not Production optimal yet due to complicated investment regulations and inefficient bureaucracy in Indonesia (Sjöholm & Lipsey, 2010;Ukhtiyani & Indartono, 2020). Furthermore, Indonesia has not optimized the benefits from the presence of FDI (Negara & Adam, 2012). Empirical studies about the nexus between FDI and the development of the Indonesian automobile industry are not yet conducted. Most studies observe the policy development of the Indonesian automobile industry (Adnan, 2014;Aswicahyono et al., 2000;Natsuda & Otsuka, 2014;Tarmidi, 2004). Some empirical studies in Indonesian automobile industry topic are already performed, such as the Total Factor Production (TFP) (Okamoto & Sjöholm, 2000) and Export fragmentation of automobile components (Soejachmoen, 2014). Nonetheless, those studies do not connect to FDI in the automobile industry. FDI has the stability inflow characteristic, the smallest risk, and more benefits compared with other investments such as portfolio investment and foreign debt since it is a longterm investment (Sagita, 2013). Hence, many previous studies investigate the FDI impact on the host country's industries. Commonly, FDI has a positive effect on productivity (Adekunle et al., 2020;Fillat & Woerz, 2011;Jude, 2012;Khan & Mehboob, 2014;Liu et al., 2016;Negara & Adam, 2012;Ni et al., 2017;Yazdan & Hossein, 2013) However, many previous studies provide various FDI effects to the host country's industries performance depending on the characteristics of the industries, the hostcountries, FDI origin countries, and the period length. FDI has positive spillovers on efficiency change but negative spillovers on technological change in the Indonesian food-processing industry, but FDI spillovers are negative in efficiency change while positive in technological progress in the Indonesian electrical machinery industry (Suyanto & Salim, 2010). Contrasting results are also shown by Suyanto et al. (2014) where FDI spillovers give negative effects on high-efficiency industries but give positive effects on low-efficiency industries in Indonesia. The time period also could cause different results where FDI backward spillovers affect TFP growth negatively in the short-term and positively in the long-term because the spillovers require technological progress stimulation (Fujimori & Sato, 2015). FDI has significant effects when it flows from developed countries to developing countries, but FDI from developing countries has no effect to host countries (Kim et al., 2015). Different industry size could experience different effect from FDI where FDI gives a negative effects on productivity for Small and Micro Enterprises (SMEs) but gives a positive effect for large enterprises (LEs) (Kurita et al., 2017). Some studies find that FDI has a low impact in developed countries (Fillat & Woerz, 2011;Yazdan & Hossein, 2013) or even is not significant (Turnbull et al., 2016;Vahter, 2011). Vahter (2011) points out that FDI spillovers has no short-term effect on productivity and may need to be materialized within a certain period.
Based on the background and various findings from previous researches, this paper attempts to investigate the determinant factors on automobile production in Indonesia particularly the FDI effect. The automobile industry is considered as a large size industry which receives numerous FDI for latest technology utilization. By combining physical, financial, and monetary variables, the empirical result is expected to give an appropriate forecasting model of automobile production in Indonesia.

RESEARCH METHODS
This paper utilizes the secondary annual data, from various sources between 1990 and 2019. Based on the data, this paper provides empirical analysis by employing Ordinary Least Square (OLS) time-series regression to observe how the trend of automobile production moved during the period. From the observation, the pattern model of data behavior can be obtained and utilized for forecasting purpose (Aljandali & Tatahi, 2018). This paper proposes the determinant factors of automobile production in Indonesia as follows:  (1) Pt is the automobile production (units) in the year t obtained from Gaikindo as the main subject of this paper. The production includes passenger vehicles and commercial vehicles such as busses and trucks. Also, this paper does not distinguish whether the production is for the domestic market or the export market.
FDIt is the investment by foreign automobile MNCs (US dollars) in the year t acquired from BKPM. As the primary predictor variable, this paper investigates the nexus between FDI and automobile production. An industry with sophisticated technology like the automobile industry requires FDI as the fundamental establishment and development (Sekuloska, 2018). This paper follows Soejachmoen's (2014) argument in which FDI inward stock is chosen as the predictor variable to avoid high fluctuation. FDIt is expected to have a positive sign.
DSt is the domestic automobile sales (units) in the year t taken from Gaikindo. The domestic sales are used to represent the automobile demand because more than 75% of the production are absorbed in the domestic market. This paper hypothesizes that DSt aff ects the production positively.
ICt is imported parts (US dollars) in the year t collected from UN Comtrade. Indonesian automobile industry still depends on imported components. Although the locally made components increase quantitatively and qualitatively, particular components still cannot be produced locally and have to be imported due to technology lackness. ICt is expected to have a positive linear relationship with the production.
ERt is the Rupiah exchange rate against the US dollar in the year t as a monetary variable downloaded from the World Bank Group. Since the Indonesian automobile industry still requires automobile components from abroad, the Rupiah depreciation might increase the production cost. As a result, the automobile demand will be lower because of a higher price. Therefore, ERt is expected to have a negative sign.
εt denotes the error term in the year t. In order to avoid a spurious regression, Gujarati & Porter (2009) suggests to carry out cointegration test by performing the stationarity test on εt. If the regression between ∆εt and εt-1 is significant, then all variables are cointegrated among them. Table 2 shows the summary result of the OLS analysis on the equation (1) and the classical assumption tests. All determinant factors are significant at 1% critical level have the similar sign as expected. The test results inform that the model has high Adjusted-R2 value, significant F-statistic, and no classical regression assumption violation. Also, the model does not generate a spurious regression because the cointegration test shows that the residuals are stationary at 1% critical level. This means that the dependent and independent variables are cointegrated. Overall, the regression model is fit to explain the automobile production in Indonesia and can be written as follows: P t = 80,692.02 + 3.6413 x 10 -5 (FDI t )+ 0.7101(DS t ) + 3.9446 x 10 -5 (IC t ) -12.4342 (ER t ) ε t ………………………………………………..(2) Prob. 0.000 Source: EViews Output Note: (*) significant at 0.1, (**) significant at 0.05, and (***) significant at 0.01 Table 2 reconfirms the findings from previous studies where FDI is a significant determinant on the large capacity industries with sophisticated technology like Indonesian automobile industry (Kurita et al., 2017;Suyanto & Salim, 2010;Zhao & Zhang, 2010). Based on the analysis, if FDI inward stock increases 1 million US dollar, then the automobile production rises about 36 units.

RESULTS AND DISCUSSION
FDI and the development of the Indonesian automobile industry cannot be separated. FDI inflows from MNCs home base to Indonesia as the host country can be realised because both sides have their interests. On one side, Indonesia has attractiveness for FDI inflow because of its large domestic market, low-cost production factors, e.g., labour wage & abundant plant sites, and friendlier tax policy. On the other side, Indonesia needs to shift its level from a developing country to industrialized country, increase employment, and receive the latest technology. Since the automobile industry utilises hi-tech machinery and sophisticated management, Indonesia has no resource and acknowledgement to create such technology and organization. Thus, technology transfer through FDI spillover is required. However, technology innovation and development require large fund and long-duration research, so MNCs are often reluctant to hand over their technology acknowledgements to the host countries. Therefore, supportive regulations from the government must be established to guarantee that MNCs are willing to transfer their technology and knowledge to local industries and technicians. Some sacrifice policies, such as tax holiday, subsidy, and/or import liberalisation, are suggested to invite more FDI and ease technology transfer. FDI inflow also has an important role for export base industry (Herlambang et al., 2016;Mahmoodi & Mahmoodi, 2016;Salsabila et al., 2015;Zysk & Śmiech, 2014). According to Kutan & Vukšić (2007), FDI brings international market links where the GoI and local industries do not have the access to the market. To increase the production scale when the domestic market is saturated, the export market can be as the main alternative to absorb excess production. Although Indonesia's automobile export has increased significantly since 2004, the competitiveness of Indonesia's automobile still weak in the export market (OECD, 2012). There are two main problems why the export market has low demand on Indonesia's automobile products.
First, the Indonesian automobile industry still utilizes manufactures with Euro2 standard while automobile importer countries have already implemented Euro4 standard. Therefore, Indonesia's automobiles with Euro2 standard are rejected, and Indonesia has to import particular components with Euro4 standard to produce automobile for the export market. This condition occurs because Indonesia is late to implement Euro4 standard in the transportation regulation since Euro4 was first announced in 2005. According to Gaikindo (2016), the regulation has been officially applied since 2018 for gasoline automobile, and it will be officially applied in 2021 for diesel automobile based on the regulation of the Minister of Environment and Forestry No. P.20/ MENLHK/ SETJEN/ KUM. 1/3/2017. Also, the GoI is still lack of Euro4 fuel supply nationally, which creates another difficulty for shifting to Euro4.
Second, Indonesia's automobile production relies too much on the Multi-Purpose Vehicle (MPV) model, while the export market requires model differentiation, particularly SUV and sedan (Gaikindo, 2020c). This condition occurs because the MPV model is the most popular and affordable for Indonesian people as the family car. As a consequence, automobile MNCs tend to fulfil the domestic market rather than produce for the export market expansion. The MPV model gained popularity when Toyota launched the Innovative International Multi-purpose Vehicle (IMV) project with Daihatsu in 2004(Toyota Motor Corporation, 2012. The project resulted in the famous low-MPV model, Toyota Avanza and Daihatsu Xenia, which is the leader of the domestic sales for more than 15 years. The IMV project also became a positive momentum for the performance of Indonesia's automobile export, where Toyota's MPV dominates the export. Based on Table 2, if the domestic demand rises ten units, then the production increases by seven units. Table 2 and Figure 2 are evidence that the domestic market strongly influences the automobile production. The domestic sales dropped severely in 1998 with only 58,000s units because of the Asian economic crisis. The best momentum for the domestic sales was in the early 2010s and reached its peak in 2013 with 1,2 million units. However, since 2015, the domestic demand seems saturated and difficult to surpass 1,5 million units. Currently, Based on (Gaikindo, 2015), the automobile production capacity is more than 2 million units per year so Indonesian automobile industry cannot rely on the domestic demand only and has to expand to the export market.
The Indonesian automobile industry Indonesia still imports the automobile components, although the local component manufacturers can supply about 80% of the automobile components for the industry (The Ministry of Industry Republic of Indonesia, 2015). The imported components are only hightech components, particularly with Euro4 standard, which cannot be produced locally. The dependency on particular imported components can make Indonesia's Current Account Deficit (CAD) worse. Table 2 informs that if the imported components increase 1 million US dollar, then the automobile production increases by 39 units. Technology transfer from FDI spillover can be used to increase the opportunity of import substitution. From Figure 3, it is clearly seen that the growth of imported components is inline with the automobile production.
Table 2 also shows that the automobile production is significantly influenced by the exchange rate US dollar to Indonesian rupiah. If the US dollar increases 1 rupiah then the Indonesian automobile industry tends to cut the production of about 12 units. The reason is that the upstream industries still need to import raw materials to produce automobile components (Gaikindo, 2020a). According to the Ministry of Industry Republic of Indonesia (2015), the upstream industries still imports raw materials such as steel, aluminium, plastic (resin), and compound rubber because the local production cannot meet the quality requirement of the automobile industry. Rupiah depreciation causes imported raw materials more expensive resulting in an increase in the production cost and the automobile price. Also, FDI has an essential role in local upstream industries development by creating a joint venture or partnership between automobile MNCs and the upstream industries.

CONCLUSION
The automobile production in the Indonesian automobile industry still does not perform optimally yet. With the production capacity of more than 2 million units, the industry only produces less than 60% of its capacity. Based on the OLS regression analysis, FDI inward stock, domestic sales, imported components, and exchange rate are significant determinants factors in automobile production.
Since the domestic market seems to be saturated, the alternative to increase automobile production is through the export market. However, the competitiveness of Indonesia's automobile is still weak because Indonesian automobile industry still utilizes Euro2 technology and relies on the MPV model for the export market. Moreover, Indonesia still highly depends on imported raw materials for automobile components production, although the local components manufacturers can provide 80% of the components.
Transfer technology through FDI spillover from automobile MNCs is required for shifting to Euro4 standard and local upstream industries improvement. Therefore, the GoI should create a friendlier policy environment to attract investment from abroad and provide better infrastructure to support Euro4 standardization.