Global and Regional Shock Transmission ― Application of Global Input-Output Table ―

This paper constructs and uses the global input-output (GIO) table with 35 industries, 29 endogenous countries and 59 exogenous countries, and develops new indices to measure the degree of shock transmission in terms of intermediate goods and value-added embodied in production induced by negative global demand shock to finished goods. After the Global Financial Crisis (GFC) in 2008, China did not experience a large decline in economic growth, even though China’s gross exports fell most severely among Asian countries. In contrast, a sharp decrease in Japanese GDP in 2009 is a consequence of a substantial decline in finished goods exports, especially in the transport equipment industry. In Japan, the shock effect tends to be absorbed in its domestic sector and is not transmitted to other foreign countries. An asymmetric pattern of shock transmission between Japan and other Asian countries can explain why Japan was more affected by GFC than other Asian countries.


Introduction
In September 2008, the world plunged into the unprecedented global financial crisis (GFC), accompanied by a deep decline in world trade. The United States experienced the most severe downturn in trade in terms of the magnitude and speed since the late 1960s (Crowley and Luo, 2011). This unprecedented collapse of world trade in 2008-09 is referred to as the "Great Trade Collapse". According to Table 1, US imports from the selected Asian and European countries fell rapidly from the 4th quarter of 2008 through the end of 2009. US imports from Japan and Malaysia became a negative growth in the 3rd quarter of 2008, and Japan appears to have been the most severely affected by the US import decline. Table 2 shows the GDP growth rate in 2009 and changes in exports 1 to the world in selected countries from 2008 to 2009. Germany and Japan (-5.6 and -5-4 percent respectively) experienced largest negative GDP growth, whereas China was affected marginally by GFC with 9.2 percent of GDP growth. In the meantime, Chinese gross export of manufacturing goods declined substantially ($207.1 billion) due to GFC compared to Japanese gross export of $176.0 billion. Therefore, it is quite unrealistic to explain the effect of GFC on GDP growth with respect to gross export shock. However, decline in finished goods export of Germany ($134.1billion), Japan ($112.1billion) and China ($72.2 billion) seems to be inter-related with respective GDP growths. Further, industry-breakdown of finished goods export shows Japan's finished goods export of transport equipment industry 2 (mainly motor vehicles) and China's electric industry 3 are by far the largest in Asia. This paper attempts to answer why Japanese GDP was worst hit by GFC and China was affected only marginally, even though China plays a central role of regional processing trade in Asia, especially in the electric machinery industry (Koopman et al., 2008, and Athukorala, 2009 1 We split gross export as sum of finished goods and intermediate goods export. See Section 3 and Appendix 5 for the details on decomposition of gross export in to intermediate goods and finished goods 2 Transport equipment industry consists 'motor vehicles' (Y18) and 'other transport equipment's' (Y19) as listed in Appendix 3. 3 Electric industry consists 'office machinery' (Y14), 'electrical machinery' (Y15), 'Radio, Television and communication equipment' (Y16) and 'optical instruments' (Y17) as listed in Appendix 3. responded to the negative demand shock by decreasing their production and supply of production goods. Such supply-side impact has often been analyzed in the literature. Ando and Kimura (2012), for instance, analyze the GFC impact on Japanese and Asian exports using the most disaggregated trade data, and decompose export changes into extensive and intensive margins to examine which factor most affected Japanese and Asian exports. However, given growing regional production network in Asia, 4 it is also important to investigate how the effect of global shock (i.e., decline in gross production) is transmitted to domestic and regional economies by tracing input sources (intermediate goods and value-added inputs) along the backward linkage. Such regional shock transmission can magnify the effect of negative world demand shock, driving regional economies into serious economic downturn. To get a clue to evaluate the progress of regional economic linkages and value chains, this paper investigates whether and how a sharp fall in finished goods exports of Asian countries to the world reduced domestic production and then induced subsequent decline in intra-Asian trade along the production chain (backward linkage) during the GFC.
One useful approach is to utilize the International Input-Output (IIO) table, where bilateral trade linkages are decomposed into two types of trade, i.e., transactions of intermediate inputs and final goods 5 at a detailed industry level. While it was generally hard to obtain the updated time series data on the IIO table, 6 in recent years numerous attempts have been made on the construction of the time-series IIO tables. 7 In particular, dissemination of World Input-Output Database (WIOD) and Organization for Economic Cooperation and Development (OECD) Inter-Country Input-Output (ICIO) Tables are expected to contribute significantly to the research on economic linkage. 8 4 Ferrarini (2013) maps global and regional linkages in production network and vertical trade, and shows a rapid increase in the degree of regional linkage in East Asia. 5 Hummels, Ishii and Yi (2001), Daudin, Rifflart and Schweisguth (2011) and Johnson and Noguera (2012). See also two special issues "Global Multiregional Input-Output Frameworks" (Ed: Dietzenbacher and Tukker, 2013) of Economic Systems Research, 25(1) and "A Comparative Evaluation of Multi-Regional Input-Output Databases" (Ed: Inomata and Owen, 2014) of Economic Systems Research,26(3) for the recent developments on internationally-linked IO tables. 8 See respective websites of the WIOD (http://www.wiod.org/index.htm) and OECD-ICIO (http://www.oecd.org/sti/ind/input-outputtables.htm) for the details of IIO tables. For research based on the WIOD data, see, for instance, Foster andStehrer (2013) andTimmer, Erumban, Los, Stehrer andde Vries (2014). Moreover, details on measuring Trade in Value Added is available on http://www.oecd.org/industry/ind/measuringtradeinvalue-addedanoecd-wtojointinitiative.htm.
In this study, we construct the Global Input-Output Table (henceforth,   YNU-GIO Table) by ourselves instead of using the WIOD and ICIO Utilizing the YNU-GIO table, we develop a new index to measure the extent of shock transmission, whereby both direct and indirect impacts of the shock can be evaluated in multiple stages of production process. To calculate the index, we conduct a simulation analysis by generating industry-specific shocks 9 to the world import demand of finished goods, which enables us to explore how and to what extent the effect of a decline in the world import demand for final goods is transmitted directly and indirectly to trade of intermediate goods and value-added through backward linkage of productions especially among Japan and Asian countries.
To anticipate the results, we show that there is an asymmetric pattern of shock transmission between Japan and other Asian countries. Japanese finished goods exports are affected substantially by the global shock, especially in the transport equipment industry, but the shock is not transmitted globally or regionally from Japan in terms of both intermediate inputs and value-added contents. In addition, the global shock that is transmitted to Japan tends to be absorbed in Japanese domestic sectors. As a manufacturing hub, China plays a major role in supplying intermediate inputs regionally and globally, especially in the electric machinery industry, which enhances the degree of regional economic integration in Asia and also inter-regional linkages between Asia, North America and Europe. However, China's value-added transmission index to the world is comparatively higher than that of Japan, implying that Japan is more 9 We use 2009 YNU-GIO table and actual change of finished goods export to the world from 2008 to 2009 to observe the effect of GFC and report the summary of results in Tables 3-5. In addition, we also use a hypothetical economic shock of 10 percent decline in finished goods export to illustrate the changes in global shock transmission patterns from 1997 to 2012 in Figure 4. Full set of results are available on request. vulnerable to regional and world demand shocks than China and other countries.
The remainder of this paper is organized as follows. Section 2 presents the methods to estimate the new indices of shock transmissionand the data construction of the YNU-GIO tables, and Sections 3 and 5 present the results of the shock transmission analysis and discuss the results. Finally, Section 5 concludes the paper.

Graphic Illustration
To evaluate the degree of global and regional economic linkages and value chains, we develop a new index of shock transmission. For a brief exposition of the new index, let us assume a four endogenous country model that consists of the United States, Japan, China and Korea (Figure 1). Suppose that the US import demand for finished goods from China declined by $17 billion, which is equivalent to a 10% decline in the actual amount of China's exports of finished goods to the United States in 2005 and regarded as a negative demand shock. As illustrated in Figure 1a We have so far assumed that only one country, China, is hit by the export shock.
But, it is usual that other countries are also affected by the global shock simultaneously.
To assess the actual pattern and impact of shock transmission, we develop a multi-country framework to estimate total induced effect in terms of intermediate goods and value-added, where all endogenous countries experience a decline of finished goods exports. In this paper, we estimate the amount of inducements in intermediate goods and value-added by assuming a change in finished goods exports to the world. However, it is possible to estimate the inducements in intermediate goods and value-added due to decline in finished goods export to a particular country such as the United States (as illustrated in the example above), or to other single country, or a region such as the North America and Europe, or the world. Once, we estimate the amount of induced effects, we calculate two types of shock transmission indices (1) shock transmitted to a country from the world and (2) shock transmitted from an individual country to the world using intermediate goods and value-added inducements. The rest of this section shows how to derive the shock transmission indices using the three-country GIO model.

Three-Country IIO Model
To evaluate the degree of shock transmission when all endogenous countries encounter a fall of finished goods exports to the world, we first calculate the change in production inducements, where finished goods exports of country 1, 2 and 3 decline by respectively. This simultaneous decline in finished goods exports induces a fall in production in three countries and can be estimated by using the global Leontief inverse matrix L as:  Figure  relative to domestic inducements. In reference to Tables 3-6, the following rule is applied to calculate shock transmission indices.

Results: Global and Regional Shock Transmission of GFC
In this section, we compute the extent of shock transmission of GFC in all endogenous countries with an actual change in finished goods export (all manufacturing industries, electric industry and transport equipment industry) to the world during 2008-2009 using the 2009 YNU-GIO Table. 17 Here, we provide two indices of shock transmissions showing that (1) to what extent of the GFC shock is transmitted to individual endogenous countries from the world (transmission to the world) and (2) to what extent individual countries transmitted its shock to the regional and global partners (transmission from the world/Asia) subject to the shock in both intermediate goods and value-added.

Shock Transmission in Intermediate goods and value-added
The upper pane of Table 3   of shock transmission to its regional partners compared to its transmission to the world.
[Insert Table 3 around here.] Interestingly, transmission of GFC shock through value-added supply and demand (value-added transmission index) provided in Table 3 illustrates that Japanese GDP experienced 13.7 percent of GFC negative shock from the world, which is by far larger than that experienced by China (9.3 percent) and other Asian countries except Germany (14.6 percent) and the US (14.3 percent). In contrast, Japan's shock transmission through value-added (8.9 percent) is much smaller than that for three major European economies (19.4 percent for Germany, and 14.3 percent each for UK and France) and China (13.9 percent). Moreover, value-added transmission indices to the world for other Asian countries are also relatively higher (ranging from 23.0 percent for Taiwan to 193.3 percent for Singapore) than Japan with an exception of 7.6 percent for Indonesia. Such situation implies that Japan is a major supplier of value-added (among Asian countries) to the world, but demands less value-added from the world compared to Germany, China and other regional partners. Or equivalently, Japanese GDP experienced substantial suffering from inbound value-added shocks due to GFC, while the shock is not transmitted (i.e., the shock is absorbed in domestic economy) to foreign and regional countries from Japan as compared to other regional countries. This is the main reason why Japanese GDP in 2009 marked lowest (-5.6 percent) growth in Asia.
Moreover, the extent of the GFC shock transmitted to Asian countries from the world and that transmitted to the world or regional partners from Asian countries illustrates asymmetric patterns of Shock transmission. For example, value-added shock transmission indices from the world are relatively smaller except for Japan and China, whereas, shock transmission to the foreign and regional economies are higher except for Japan, China and Indonesia.

Shock Transmission in Electric Machinery and Transport Equipment
Tables 4  [Insert Table 4 Table 5 presents the effect of GFC in transport equipment industry. In contrast 18 We select two industries (electric machinery and transport equipment) to discuss the results , because these two are major industries to promote production networks and economic integration in Asia 19 The weight of GFC shock by industry is calculated as ratio of industry-specific direct finished goods export shock to manufacturing industry's direct finished goods export shock. to electric industry, the weight of GFC shock is 54.3 percent in Japanese transport industry (= -61.0/-112.2, see Tables 3 and 5 Tables 3 and 5), however, the magnitude of shock transmitted to Korea from the world is relatively low and that transmitted to the world from Korea is high. Such evidence indicates that GFC predominantly affected the Japanese economy due to largest decline of finished transport equipment goods export compared to the other Asian countries. Moreover, Tables 4 and 5 evidently illustrates asymmetric shock transmission patterns not only across the two selected industries, but across countries in Asia also.

Discussions: Changes in Regional Linkages
We have so far analyzed the effect of GFC on shock transmission focusing on the Asian region using 2008-2009 data. It is also worth investigating how regional and global linkages for intermediate inputs and value-added contents changed. We first, calculate and summarize transmission of hypothetical shock using 10 percent decline of finished goods export to the world in each year from 1997 to 2012 and corresponding year's YNU-GIO table in terms of regional procurements of intermediate goods and value-added from Asia, North America and Europe. Figure 4 shows not only the intraand inter-regional linkage of procurements of intermediate inputs and value-added contents but it also shows the world's dependence on each region. In the far left figure, Asian countries exhibit a remarkable growth in regional procurements of intermediate inputs and value-added from 1997 to 2006, however the Asian intra-regional dependence significantly declined until 2009 and started to gain some momentum later.
On the other hand, Asia's procurements from North America and Europe do not show an increase during 1997-2012 period in general. In particular, growing trend in regional procurement of value-added demonstrated by Asia supports the recent findings that regional economic integration in Asia has been mainly driven by growing regional production network and fragmentation. 20 The second third left figures in Figure 4 indicate that the level of regional procurements of intermediate inputs and value-added is higher in Europe than in Asia and North America. An increasing trend in European intra-regional dependence, subject to both procurements, can be observed during the sample period, whereas North value-added terms, Europe is still the major supplier to the world, however, the gap between Asia and Europe has reduced significantly. The above observation suggests that Asia shows the significant progress of regional economic linkages and also becomes more integrated into global production network through intermediate input and value-added trade.

Concluding Remarks
We developed a new index to measure the extent of shock transmission due to global demand shock in finished goods export. The finished goods demand shock transmits to respective country's domestic production, which in turn transmits to its domestic economy and foreign partner economies via induced intermediate goods and value-added. We standardize the foreign inducements relative to domestic inducements to make an international comparison. Our methodology differs from the GVC (Global Value Chain) approach as the GVC model treats intermediate goods trade as both endogenous and exogenous variables simultaneously. Moreover, we use an entirely new set of sixteen annual globally linked IO tables, called as YNU-GIO tables, from 1997 to 2012 to reveal the extent of shock transmission due to exogenous shock in finished goods trade. The new data set focuses primarily on dynamic structural changes and economic integration in Asian region that are not covered fully by existing data sets.
We have shown that there is an asymmetric pattern of shock transmission between Japan, China, Korea and other Asian countries. Japan is affected substantially by the direct global shock (specifically in transport equipment industry) and the induced shock is not transmitted to foreign countries from Japan as compares to the shock transmitted to foreign countries from other Asian and non-regional countries in terms of both intermediate inputs and value-added contents. The global shock tends to be absorbed in the Japanese domestic sectors. As a manufacturing hub, China plays a major role in supplying intermediate inputs regionally and globally, especially in the electric machinery industry. However, China's rate of shock transmission to foreign countries is higher than the Japanese one (in particular, value-added inducements in foreign countries), which means that the Chinese GDP sector experiences only a marginal negative shock. Even though China enhances regional economic integration through intermediate input transactions, Japan is more vulnerable to regional or global demand shock.
Our research can be extended in the following ways. Processing trade accounts for a large share in China's total trade, but it is not taken into account in this paper.

Appendix 1. GVC model under the Three-Country IIO Model
Let us assume a three-country GIO table presented in Figure A1, where each country produces in a single tradable sector. Each country produces a good that can be consumed as a final good or used as an intermediate input. 22 is a vector of gross output and is vector of value-added inputs. Then we can easily derive the Input-Output equation, for the GIO table given in Figure A1, in matrix form as    22 We assume that each country has only one production sector. This assumption can be easily extended to a multi-production sector model with the same matrix and vector notations.
Substituting equation (3) in equation (1), we get, Equation (5) implies that the gross output vector (Y) is sum of productions induced by domestic consumption (F D ) and export (E F ) vectors of finished goods. Thus the value-added associated with induced production can be written as where, v is diagonal matrix of value-added coefficients.
The global value chain (GVC) in existing literatures estimates the induced value-added generated by gross export (say, ) using the GIO data as follows Here, equation (7) represents the value-added associated with the production induced by export of intermediate goods ( Z LE ) and that induced by export of final goods ( F LE ). Now, the gross production under GVC assumption (say, YGVC) can be written as sum of productions induced by domestic final demand vector F D , export vector of final goods E F and export vector of intermediate goods E Z , and using equation (5) we get, Note that, for the given global Leontief inverse matrix L and non-zero E Z , YGVC exceeds the gross output Y provided in the GIO  Table   (

Appendix 3: List of production industries of the YNU-GIO table
Industry Code Name of Industry

Y01
Agriculture, hunting, forestry and fishing Y02 Mining and quarrying Y03 Food products, beverages and tobacco Y04 Textiles, textile products, leather and footwear Y05 Wood and products of wood and cork Y06 Pulp, paper, paper products, printing and publishing Y07 Coke, refined petroleum products and nuclear fuel Y08 Chemicals and pharmaceuticals Y09 Rubber and plastics products Y10 Other non-metallic mineral products Y11 Basic metals Y12 Fabricated metal products, except machinery and equipment Y13 Machinery and equipment Y14 Office, accounting and computing machinery Y15 Electrical machinery and apparatus Y16 Radio, television and communication equipment Y17 Medical, precision and optical instruments Y18 Motor vehicles, trailers and semi-trailers Y19 Other transport equipment Y20 Other Manufacturing; recycling (include Furniture) Y21 Electricity, Gas and Water supply Y22 Construction Y23 Wholesale and retail trade; repairs Y24 Hotels and restaurants Y25 Transport Y26 Post and telecommunications Y27 Finance and insurance Y28 Real estate activities Y29 Renting of machinery and equipment Y30 Computer and related activities Y31 Research and development Y32 Other Business Activities Y33 Public administration and defense; compulsory social security Y34 Education Y35 Health, social work and other services final goods in all the endogenous countries, which is finally used to reorganize the globally linked Input-Output table.

Appendix 5: Estimation of the YNU-GIO Tables
Our estimation process of annual YNU-GIO tables follows flow-chart presented in Figure A2. First of all, we collect NIOTs, 1 for 29 endogenous countries, valued in corresponding national currencies and rearrange them according to the YNU-GIO