Sector-Specific and Spatial-Specific Multipliers in Indonesian Economy: World Input-Output Analysis

This article discusses on sectoral-specific and spatial-specific multipliers in Indonesian economy using 6-country-30 sector input-output tables for the year 2000, 2005, 2010 and 2014. The result shows that firstly, in all years, there were 20 sectors with total output multipliers more than 2. Flow-on effects were higher than initial effects. These sectors should be prioritized if increasing of total output is the objective of Indonesian economic development as total output will be created with less intial efforts. Secondly, in the year of 2000, average percentage of multipliers occurred in own-sector was 56.23 per cent, and increase slightly in 2005 (57.38%) dan 2010 (58.93%), but decrease in 2014 (57.98%). Correlation between total output multipliers and percentage of multipliers occurred in other-sector was positive and very strong. The higher total output multipliers, the higher percentage of multipliers occurred in other-sector. Thirdly, in the year of 2000, average percentage of multipliers occurred in other-countries was 21.34 per cent and decrease slightly in 2005 (20.22%) and 2010 (18.14%), but increase in 2014 (20.55%). Correlation between total output multipliers and percentage of multipliers occurred in other-countries were positive and very strong. The higher total output multipliers, the higher multipliers occurred in other-countries.


Indonesia is the largest economy in Southeast
Asia and is one of the emerging market economies of the world. This archipelago comprising five main island (Muchdie, 2011) and more than 17.000 small island (BPS, 2014), more than 250 million people lived in 34 provinces. The country is also a member of G-20 major economies and classified as a newly industrialized country (World Bank, 2017). It is the sixteenth largest economy in the world by nominal GDP and is the seventh largest in terms of GDP (PPP). Indonesia still depends on domestic market, and government budget spending and its ownership of state-owned enterprises and the administration of prices of a range of basic goods including, rice, and electricity plays a significant role in Indonesia market economy, but since the 1990s, the majority of the economy has been controlled by private Indonesians and foreign companies (Adhi, 2015).
In macroeconomics, a multiplier measures how much an endogenous variable changes in response to a change in some exogenous variables (Pindyck & Rubinfeld, 2012). In monetary microeconomics and banking, the money multiplier measures how much the money supply increases in response to a change in the monetary base (Krugman & Wells, 2009;Mankiw, 2008). Multipliers can be calculated to analyze the effects of fiscal policy, or other exogenous changes in spending, on aggregate output. Other types of fiscal multipliers can also be calculated, like multipliers that describe the effects of changing taxes. Literature on the calculation of Keynesian multipliers traces back to Kahn's description of an employment multiplier for government expenditure during a period of high unemployment. At this early stage, Kahn's calculations recognize the importance of supply constraints and possible increases in the general price level resulting from additional spending in the national economy (Ahiakpor, 2000). Hall (2009) discusses the way that behavioral assumptions about employment and spending affect econometrically estimated Keynesian multipliers.
The literature on the calculation of I-O multipliers traces back to Leontief economic IO model in 1930s to the broadening spectrum of applications over the years is well documented (Hewing & Jensen, 1987;Rose & Miernyk, 1989;Lahr, 1993;Dietzenbacher & Lahr, 2004;Bjerkholt &Kurz, 2006;Miller &Blair, 2009). The diversity of theme reflects the blossoming of input-ouput as a key analytical framework in the field of economic systems research and its sub-diciplines, with relevance to numerous reseach fields (Suh & Kagawa, 2005;Timmer and Aulin-Ahmavaara, 2007;Debresson, 2008;Wiedmann, 2009;Los & Steenge, 2010, Duarte & Yang, 2011Tukker & Dietzenbacher, 2013;Inomata & Owen, 2014;Okuyama & Santos, 2014). In addition, inputoutput has established itself as an invaluable prcticl tool widely used by governments, industry and other national and international organisation (Meng et al., 2013;OECD, 2015;United Nations, 2017). Richardson (1985) notes the growth of survey-based regional input-output models in the 1960s and 1970s that allowed for more accurate estimation of local spending, though at a large cost in terms of resources. To bridge the gap between resource intensive survey-based multipliers and "off-the-shelf" multipliers, Beemiller (1990) describes the use of primary data to improve the accuracy of regional multipliers. The literature on the use and misuse of regional multipliers and models is extensive. Coughlin & Mandelbaum (1991) provide an accessible introduction to regional I-O multipliers. They note that key limitations of regional I-O multipliers include the accuracy of leakage measures, the emphasis on short-term effects, the absence of supply constraints, and the inability to fully capture interregional feedback effects. Grady & Muller (1988) argued that regional I-O models that include household spending should not be used and argue that cost-benefit analysis is the most appropriate tool for analyzing the benefits of particular programs. Mills (1993) noted the lack of budget constraints for governments and no role for government debt in regional IO models. As a result, in less than careful hands, regional I-O models can be interpreted to over-estimate the economic benefit of government spending projects. Hughes (2003) discussed the limitations of the application of multipliers and provides a checklist to consider when conducting regional impact studies. Application of regional multipliers in the context of tourism impact studies, one area where the multipliers are commonly misused have been discussed by Harris (1997). Siegfried, et al. (2006) discussed the application of regional multipliers in the context of college and university impact studies.
Input-output analysis, also known as the inter-industry analysis. The fundamental purpose of the input-output framework is to analyze the interdependence of industries in an economy through market based transactions. Input-output analysis can provide important and timely information on the interrelationships in a regional economy and the impacts of changes on that economy. Unlike singleregion input-output model, global multi-regional input-output tables are able to shed light on the complex interdependencies in a globalised world, such as outsourcing of productionand associated environmental impact, vertical specilisation or value added embedded in trade and global value chain (Hertwich & Peters, 2009;Hummels et al., 2001;Lenzen et al., 2012;Wiebe et al., 2012;Timmer et al., 2014. The objective of this paper is to calculates, presents and discusses on sectoral and spatial multipliers in Indonesian economy using 6-country-30sector input-output tables for the year 2000, 2005, 2010 and 2014 processed from World Input-Output Tables.

Method of Analysis
An input-output table records the "flows of products from each industrial sector considered as a producer to each of the sectors considered as consumers" (Miller & Blair, 2009). It is an "excellent descriptive device" and a powerful analytical technique (Jensen et al., 1979). In the production process, each of these industries uses products that were produced by other industries and produces outputs that will be consumed by final users (for private consumption, government consumption, investment and exports) and also by other industries, as inputs for intermediate consumption (Oosterhaven & Stelder, 2007;Timmer et al., 2015). These transactions may be arrayed in an input-output table, as illustrated in Table 1.
The columns of Table 1 provide information on the input composition of the total supply of each product j (X j ), this is comprised by the national production and also by imported products.  The input-output interconnections illustrated in Table 1 can be translated analytically into accounting identities. On the supply perspective, if X ij denote the intermediate use of product i by industry j and y i denote the final use of product i, we may write, to each of the n products: A X i = AA X ij + BA X ij + CA X ij + … + ZA X ij + A VA i (1) On the demand side, we know that: in which w j stands for value added in the production of j and m j for total imports of product j. It is required that, for i = j, x i = x j , i.e., for one specific product, the total output obtained in the use or demand perspective must equal the total output achieved by the supply perspective. These two equations can be easily related to the National Accounts' identities. In general term, equation (1) can be written as: National Input-Output  (Timmer et al., 2016;2015). Calculation of total and disaggregated multipliers, sector-specific multipliers and country-specific multipliers were following West (1990) and modified formula of DiPasquale & Polenske (1980). West (1990) defined the major categories of response as: initial, first-round, industrial-support, consumptioninduced, total and flow-on effects. Total effect is calculated as summation of initial, direct-effect (first-round), indirect-effect (industrial-support) and consumption induced effect (as matrix is closed to house-hold row and column, which was not calculated in this study). Flow-on effect is defined as the different between total and initial effects. Modified from DiPasquale & Polenske (1980), sector-specific multipliers of output are calculated as ∑ cr b ij; c = 1..., n, and country-specific multipliers of output is calculate as ∑ cs b ij; i = A,.., Z. Note that c and r are the m origin and destination countries, i and j are the n producing and purchasing sectors, cr b ij is the element of inverse of Leontief matrix, m is the number of country and n is the number of sectors. Sector classifications is available in Appendices. In the year 2000, 20 sectors had total output multipliers more than 2. It means that the sum of direct and indirect effects were more than the initial effects; the flow-on effects was higher than initial effect. These sectors were: Sector-5 (2.1886), Sector-6 (2.4223), Sector-7 (2.1122), Sector-8 (2.5858), Sector-9 (2.3655), Sector-11 (2.2409), Sector-12 (2.4163), Sector-13 (2.4556), Sector-14 (2.0958), Sector-15 (2.4848), Sector-16 (2.5590), Sector-17 (2.5090), Sector-18 (2.4984), Sector-19 (2.5852, Sector-20 (2.2899), Sector-21 (2.6507), Sector-22 (2.4018), Sector-24 (2.2833), Sector-25 (2.4052), and Sector-27 (2.1748). Based-on direct and indirect effects of change in final demand, these 20 sectors had more than 20 per cent of direct effects and another 20 per cent of indirect effects, made totally more than 40 per cent of total effects. It means that to increase 1 unit of total output, it only needs to increase final demand (export, for instance) by less than 50 per cent. Other 10 sectors had total multipliers less than 2; meaning that increasing 1 unit of final demand would increase total output by less than 2. Flow-on effects created were less than 1 unit.

Figure-2: Disaggregated Output Multipliers in Indonesian Economy: 2010 and 2014
Source: Processed from WIOT, 2017 In the year 2014, 20 sectors that had total ). These 20 sectors had more than 20 per cent of direct effects and another 20 per cent of indirect effects, made totally more than 40 per cent of total effects. It means that to increase 1 unit of total output, it only needs to increase final demand (export, for instance) by less than 50 per cent. Other 10 sectors had total multipliers less than 2; meaning that increasing 1 unit of final demand would increase total output by less than 2. Flow-on effects created were less than initial effects.

Sector-Specific Multipliers
Sector-specific multipliers separate multipliers that occurred in own-sector and that occurred in other sectors. it could be shown that in all sectors multipliers occurred in own-sector were higher that initial effect. It means that other effects of multipliers, for instance the direct-effect, might also be occurred in own-sector. Take for example Sector-1, multiplier occurred in own-sector was 61.29 per cent, meanwhile initial effects was 59.56 per cent. Off course, all initial effects took-place in ownsector. The rest of multipliers occurred in ownsector (7.27 %) might by direct effect (15.57%). In other case, multipliers occurred in own-sector might be consist of initial effect and direct effect. In Sector-2, for instance, multiplier occurred in own-sector was 96.68 per cent, meanwhile initial effect was 71.94 per cent. The rest of multiplier occurred in own-sector (24.74%) might be partly by direct effect (14.43%) and or partly by indirect effect (13.63%). Regression analysis showed that correlation between multiplier occurred in ownsector and the initial effect of multiplier was very strong (0.94) and regression coefficient was statistically significant as calculated t-statistic (14.009) was higher than critical-value of t-distribution with n-1=29 (t-table= 1.699 at α=5% or 2.045 at α= 2.5%).
In Examining initial effect and multipliers that occur in own-sector, the question was whether all initial effects occurred in own-sector. Might it be possible that direct and indirect effects also occurred in own-sector? As all sectors experiencing that multipliers occurred in ownsectors were higher than initial effects, it can easily concluded that all initial effects occurred in own-sector. Of course, it would be possible if direct effects, partly or fully, were also occurred in own-sector. For example, in Sector-1 different between multiplier occurred in own-sector and initial effect of multiplier was 5.28 per cent; meanwhile the percentage of direct effect was 13.45 per cent meaning that direct effect partly occurred (5.28%) in own-sector, other part of direct effect (8.17%) occurred in other-sector. In this case, it seems impossible that indirect effect of multiplier occurred in own-sector. Regression analysis shown that correlation between multiplier occurred in own-sector and the initial effect of multiplier was very strong (0.93) and regression coefficient was statistically significant as calculated t-statistic (8.939) was higher than critical-value of t-distribution with n-1=29 (t-table= 1.699 at α=5% or 2.045 at α= 2.5%).  Sector In the year of 2014, average percentage of total multiplier occurred in own-sector was 57.98 per cent; decreasing from that at the year of 2010 (58.93%). In this year, there were 22 sectors had multiplier more than 50 per cent occurred in own-sector, namely: Sector-1 (78.48%), Sector-2 (77.37%), Sector-3 (79.11%), Sector-4 (76.46%), Sector-5 (56.97%), Sector-6 (55.60%), Sector-7 (53.11%), Sector-8 (56.07%), Sector-10 (55.37%),  Sector-30 (74.23%). All initial effects occurred in own-sectors. Direct effect of multipliers might be occurred in own-sector, indirect effect might be not. Regression analysis shown that correlation between multiplier occurred in own-sector and the initial effect of multiplier was very strong (0.87) and regression coefficient was statistically significant as calculated t-statistic (9.217) was higher than critical-value of t-distribution with n-1=29 (t-table= 1.699 at α=5% or 2.045 at α= 2.5%).  All initial effects of multipliers occurred in own-country, but direct and indirect effects might be occurred in other countries. Analyzing import components in production would be very useful in explaining this situation. Regression analysis between multipliers occurred in own-country and initial effect of multipliers shown that correlation between them was very strong (0.85) and regression coefficients (0.53) was statistically significant as calculated t-statistic (8.547) was higher than critical-value of t-distribution with n-1=29 (t-table= 1.699 at α=5% or 2.045 at α= 2.5%).

Spatial-Specific Multipliers
In All initial effects of multipliers occurred in own-country, but direct and indirect effects, partly or fully, might be occurred in other countries. Regression analysis between multipliers occurred in own-country and initial effect of multipliers shown that correlation between them was very strong (0.81) and regression coefficients (0.49) was statistically significant as calculated t-statistic (7.183) was higher than critical-value of t-distribution with n-1=29 (t-table= 1.699 at α=5% or 2.045 at α= 2.5%).
In the year of 2010, average percentage of multipliers occurred in own-country was 81.86 per cent; 18.14 per cent occurred in other countries, in Australia (0.50%), China (2.88%), Japan (1.85%), the USA (0.84%) and the rest of the World (12.07%).

Discussions
This section discusses three important findings. Firstly, total output multipliers disaggregated into initial, direct, indirect and total effects. Flow-on effect is the different between total effect and initial effect; or flowon effect is the summation of direct effect and indirect effects. In all years of study, 2000,2005,2010 and 2014, there were 20 sectors had total multipliers of more than 2. Compared to study by Muchdie & Kusmawan (2018), Indonesian had more sectors in total output multipliers that more than 2 compared to that of the United States. There were two specific findings from total output multipliers in Indonesian economy. Firstly, all sectors with total output multipliers more than 2 had initial effects less than 50 per cent. Flow-on effects were higher than initial effects. It means that direct and indirect effects of change in final demand were higher than initial effects. These sectors should be prioritized if increasing of total output is the objective of Indonesian economic development. Total output will be created with less initial efforts. Secondly, all sectors with total output multipliers less than 2 had initial effects more than 50 per cent. Flow-on effects were less than initial effects; direct and indirect effects were less than initial effects. In other word, initial effects were higher than direct and indirect effects. If these sectors were chosen as priority sectors then more initial efforts will be needed to increase total output. This priority setting will be inappropriate as it is not the most efficient way in producing output.
Secondly, related to sector-specific multipliers, in the year of 2000, average percentage of multipliers occurred in own-sector was 56.23 per cent, and increase slightly in 2005 (57.38%) and 2010 (58.93%), but decrease in 2014 (57.98%). Study by Muchdie and Kusmawan (2018)  were higher than critical value of t-distribution with n-1=29 (t-  with total output multipliers more than 2 had less than 80 per cent of total multipliers occurred in own-country; more than 20 per cent of total multipliers occurred in other-countries. Secondly, all sectors with total output multipliers less than 2 had more than 80 per cent of total multipliers occurred in own-country; less than 20 per cent occurred in other countries. This finding implies that the higher total output multipliers, the higher multipliers occurred in other-countries.

Conclusions
From important findings as discussed in the last the Section, some conclusions could be drawn. Firstly, total output multipliers were mostly determined by direct and indirect effects of increase in final demand. In all years, there were 20 sectors with total output multipliers more than 2 meaning that flow-on effects were higher than initial effects. Direct and indirect effects of change in final demand were higher than initial effects.
These sectors should be prioritized if increasing of total output or increasing GDP is the objective of Indonesian economic development. Total output will be created with less initial efforts. Secondly, in the year of 2000, average percentage of multipliers occurred in own-sector was 56.23 per cent, and increase slightly in 2005 (57.38%) and 2010 (58.93%), but decrease in 2014 (57.98%). Correlation between total output multipliers and percentage of multipliers occurred in othersector was positive, very strong and statistically significant. The higher total output multipliers, the higher percentage of multipliers occurred in other-sector. Thirdly, in the year of 2000, average percentage of multipliers occurred in othercountries was 21.34 per cent and decrease slightly in 2005 (20.22%) and 2010 (18.14%), but increase in 2014 (20.55%). Correlation between total output multipliers and percentage of multipliers occurred in other-countries were positive, very strong and statistically significant. The higher total output multipliers, the higher multipliers occurred in other-countries.

Acknowledgement
Authors thank to Director of Graduate School and Dean of Faculty of Economics and Business, Universitas Muhammadiyah Prof Dr. HAMKA (UHAMKA) for their support and encouragment. Lemlitbang (UHAMKA) for faciliting instrument to do some calculations. The comments by the anonymous referees have also been extremely useful for improving his paper. Adhi, A. (2015). 80 Persen Industri Indonesia