# A comparative study on industrial spillover effects among Korea, China, the USA, Germany and Japan

Yong-Ki Min (Business School, Sogang University, Seoul, Korea)
Sang-Gun Lee (Business School, Sogang University, Seoul, Korea)
Yaichi Aoshima (Institute of Innovation Research, Hitotsubashi University, Tokyo, Japan)

ISSN: 0263-5577

Publication date: 8 April 2019

## Abstract

### Purpose

Starting from industry 4.0 in Germany and followed by the New Strategy for American Innovation in the USA and the smartization strategy in Japan, developed countries are pushing nation-wide innovation strategies. Similarly, China is pursuing the Made in China 2025, and Korea announced the Manufacturing Industry Innovation 3.0 strategy. However, few researchers have identified the industrial structure that establishes the foundation of the 4th Industrial Revolution or have derived strengths and weaknesses to provide implications on policy formulation through quantitative comparison with developed countries. Therefore, the purpose of this paper is to analyze the spillover effect of the information and communication technology (ICT) industry (the foundation of the 4th Industrial Revolution) and machinery·equipment industry (the foundation of smart manufacturing through convergence with ICT industry).

### Design/methodology/approach

This study examines the industrial spillover effects of the ICT industry and machinery·equipment industry in the USA, Germany, Japan, China and Korea by using the World Input–Output Table from 2000 to 2014.

### Findings

The results showed that backward linkage effect of the ICT Industry are high in the order of Korea≑China>Japan>the USA≑Germany, and forward linkage effect of the ICT industry are high in the order of Japan ≑> the USA≑Korea ≑> China ≑> Germany. Backward linkage effects of the machinery·equipment industry are high in the order of China>Japan≑Korea>the USA>Germany, and forward linkage effects of the machinery·equipment industry are high in the order of China>Korea>Germany≑Japan≑the USA.

### Practical implications

China and Korea encourage active government investment in ICT and machinery·equipment industries, especially the intentional convergence between ICT and machinery·equipment industries is expected be generate higher synergy. The “innovation in manufacturing” strategy in the USA that utilizes its strength in ICT services seems appropriate, whereas Germany needs to revitalize the ICT industry to strengthen its manufacturing industry. Japan’s strategy is to focus its ICT capabilities on robot sector. While the scope of innovation is limited, its synergy is worth expecting.

### Originality/value

This study attempted to provide a theoretical approach to the determination of national policy strategies and provide practical implications for response to the impacts of the 4th Industrial Revolution, by comparing the inducement effects of ICT and machinery·equipment industries between major countries.

## Citation

Min, Y.-K., Lee, S.-G. and Aoshima, Y. (2019), "A comparative study on industrial spillover effects among Korea, China, the USA, Germany and Japan", Industrial Management & Data Systems, Vol. 119 No. 3, pp. 454-472. https://doi.org/10.1108/IMDS-05-2018-0215

## Publisher

:

Emerald Publishing Limited

## 1. Introduction

Industry 4.0 was first used in the 2011 Hannover Expo, and assigned as a core future project for the 4th Industrial Revolution by German Government in 2012 (Drath and Horch, 2014). During the opening ceremony of the 2015 Hannover Exhibition Ground, Angela Merkel, the Chancellor of Germany, emphasized integrated industry, as well as the necessity of integrating all production processes and the close cooperation with information and communication technology (ICT) and machinery industries. In addition, Klaus Schwab, the Founder and Executive Chairman of the World Economic Forum, stated the arrival of the 4th Industrial Generation, which is distinctly different from the 3rd Industrial Generation (Schwab, 2016). Although some people argue that it is too early to call the development 4th Industrial Generation, we are reaching an age of not only digitalization but also “digital transition,” where the entire society utilizes digital technologies (OECD, 2017).

To respond to such changes, countries including the USA, Germany and Japan are promoting a “nation-wide innovation” strategy that utilizes their individual strengths. The USA developed “A Strategy for American Innovation” for creating new jobs and strengthening ICT industry leadership, and promoted advanced manufacturing partnership for reviving the manufacturing industry based on the private sectors’ capability of Big Data, IoT, etc. Germany contributes to “Industry 4.0,” the industrial transformation with automation, by converging general machinery and outstanding labor in manufacturing, in order to maintain its lead in manufacturing. Japan is the first country that set “the 4th Industrial Revolution” as a national strategy, actively utilizing its strong robotics technology in order to increase industrial competitiveness and accelerate the socio-economic system. Similarly, China is promoting the “Made in China 2025 Plan” by benchmarking Germany’s Industry 4.0 in its 13th five-year plan. Korea enforces various strategies such as the “Mid- to Long-Term Master Plan in Preparation for the Intelligent Information Society” and “Manufacturing Innovation 3.0.”

The cyber-physical system (CPS) is the foundation of the 4th Industrial Revolution and refers to the convergence between ICT and other industry (Schwab, 2016), which can increase the competitiveness of manufacturing industry by optimizing the production process. ICT, machinery·equipment and biomedical industries are directly related to the core technologies of 4th Industrial Revolution, including AI, IoT, Big Data, automation and sensors. Many studies and policies are available at present, when it is necessary to develop active strategies that utilize the 4th Industrial Revolution as a new growth and leap. However, few studies have quantitatively compared the infrastructure of the 4th Industrial Revolution with the “nation-wide innovative” countries that lead the 4th Industrial Revolution, and the corresponding strategic directions.

Therefore, this study aims to use input–output (IO) analysis to analyze the spillover effect of the ICT industry (the foundation of the 4th Industrial Revolution) and the machinery·equipment industry (the foundation of smart manufacturing through convergence with ICT industry), by using the World Input–Output Table (WIOT) during 2000–2014. By quantitatively comparing the forward and backward linkage effects of the ICT industry and machinery·equipment industry between China, Korea and the three countries that are promoting nation-wide innovation – the USA, Germany and Japan – this study examines each country’s global strengths and weaknesses in managing the 4th Industrial Revolution and suggests national innovation strategies for each country.

## 2. Literature review

### 2.1 Arrival of the 4th Industrial Revolution

Although Jeremy Rifkin and Rober Gordon argue that the statement by Klaus Schwab (2016) on the arrival of the 4th Industrial Revolution from the perspectives of speed, scope and system impact is untimely and inappropriate. But all countries are accelerating innovation to respond to the new world with different management of society and human role.

Schwab (2016) defined the 4th Industrial Revolution as CPS. CPS was first suggested by the USA in the mid-2000s, but Germany is leading in manufacturing innovation by Industry 4.0 (Drath and Horch, 2014).

The most noticeable point of the 4th Industrial Revolution is “manufacturing innovation,” which uses the ICT technology to automate the production process and intelligently implement inter-process communication systems, in order to achieve “digitalization of manufacturing process” and “servitization of products,” leading to innovative changes including customized small-quantity production and IoT-based customer service. Germany’s “Industry 4.0” is a representative policy of such changes. At the enterprise level, the industrial internet model of GE can be a reference. OECD focuses on not only digitization but also “digitalization/digital transformation,” where digital technology is utilized over the whole community, demonstrating that the widespread economic and social changes along with technology revolution will make a huge difference to our lifestyle (OECD, 2017).

As presented in Table I, most existing studies on the 4th Industrial Revolution have focused on employment and manufacturing innovation.

### 2.2 ICT industry and machinery·equipment industry

ICT industry is the driving force of economic and social growth (Fransman, 2009, and has a strong impact on all industries because it grows rapidly and induces innovation of enterprises. ICT industry has a higher forward linkage than backward linkage (Mattioli and Lamonica, 2013); in particular, telecommunication industry is a core industry on the supply side with its high forward linkage (Garcia and Vincente, 2014). In the USA, ICT manufacturing industries took up 9.5 percent of the whole manufacturing from 2011 to 2014, which was lower than that between 1981 and 1985, while the revolution of ICT service industry runs better than that in Europe and other developing countries (Fransman, 2009). The percentage of ICT manufacturing in Germany decreased the most from 13.0 to 9.6 percent, whereas that of ICT manufacturing in Japan increased from 12.3 to 15.7 percent. China, the world’s largest exporter of ICT goods, has been mainly relying on quantitative growth of HW based on its manufacturing technology as a global manufacturer. Along with the globalization of manufacturing companies in China, internet companies in China start expanding their impacts in the global market after steady growth in the domestic market. That is, the traditional HW-oriented industrial structure is changing to a new SW- and internet-oriented industrial structure. On the other hand, Korea has an export-oriented structure in ICT manufacturing, which held 22.2 percent of the whole manufacturing industry between 2011 and 2014. On the other hand, the growth in ICT service is slow, and foreign direct investment increased rapidly by 46.9 percent from 2012 to 2015, while capital investment decreased (Lee, 2017).

Machinery·equipment industry, including the manufacturing of machine tools, industrial robots, automation, mechanical elements and equipment, and each industry’s specialized machines tools, is a key industry that provides the facilities infrastructure for manufacturing industry, such as automation, new materials and sensor technology development (Arnold, 2001). Moreover, machinery industry is a key industry for implementing smart factory, which has high linkage effects, especially backward linkage effect, on other industries (Kwak and Park, 2009). Machinery industry is one of the top five manufacturing industries in the USA in 1984, occupying 6.4 percent of the total manufacturing outputs, but it then disappeared without a trace during 2011–2014. In contrast, machinery secured second place among all domestic manufacturing industries in Germany, and its proportion increased from 12.8 percent in 1984 to 15.2 percent in 2014. The proportion of machinery to all domestic manufacturing industries in Japan increased from 8.2 to 10.4 percent, moving up from the fifth to third place in 2014. Since the Chinese Government promotes the “Zizhu chuangxin(自主創新)” policy in its 11th five-year plan in 2006, it implemented policies to foster general industrial machinery in earnest. Machinery production has been expanding rapidly until 2011 and then slowed down, but it remains as the absolute no. 1 producer and importer of the world. As of 2014, it accounted for 11.5 percent of the global exports. Korea’s machinery industry accounts for 7.42 percent of the total manufacturing industry, takes the fifth place in production amount, third place in number of businesses and second place in number of employees. Korea’s machinery·equipment industry has high proportions of small businesses and employees, and thus is a very important industry from not only the economic perspective but also the employment perspective (Statistics Korea, 2017) (Table II).

The manufacturing industry was reinvented as the central axis of the twenty-first century economic growth and job creation after losing its leading position to the service industry. The USA desired for “manufacturing renaissance” by constructing manufacturing innovative networks, accelerating digitalization and establishing R&D centers. On the other hand, Germany, a manufacturing powerhouse based on its traditional machinery industry, started promoting “Industry 4.0” strategy to “construct complete automation production systems and optimize all manufacturing productions” in earnest since 2013. Similarly, Japan aims to transform to the most advanced economic and social system by using its strength, such as robotics, IoT, Big Data and AI. China promoted the “Made in China 2025 Plan” and mapped out a specific plan in May 2015 to boost manufacturing industry. This is a key strategy of the Chinese Government to pursue industrial advancement, and China is striving to become a manufacturing powerhouse across three stages over the next 30 years. In Korea, because the rate of increase of ICT convergence is the steepest within machinery industry, policy contributions are made to establish Smart Factory through convergence between ICT and machinery (Song and Kwak, 2012). As of 2012, the USA, China, Germany, Japan and Korea account for 48 percent of the global GDP and 35 percent of export, and the economies are constantly expanding. The five countries are leading the 4th Industrial Revolution, and each of them is promoting the convergence between ICT and machinery equipment industries for smart Society and smart manufacturing. Therefore, it is meaningful to examine the inducement effect of ICT and machinery·equipment industries of each of the country, and provide implications on their national strategic directions.

### 2.3 World Input–Output Tables

WIOT is the output of the World Input–Output Database (WIOD) project, which is organized by scholars and professionals from 11 organizations and sponsored by European Commission. This project, published in May 2009, reports IO tables of 27 EU countries and 13 other major countries in the world, and is the first inter-industry relation table based on other national account statistics. Subsequently, WIOD released a revised edition in November 2016, which provides a time-series IO table from 2000 to 2014 of EU country and major countries. WIOT uses the supply-use tables of individual countries as basic data, but it also uses both international trade data and socio-economic or environmental data composed based on industrial standards to compile statistics more intuitively and elaborately than directly linking IO tables of each country. In addition, this study used the frequently updated statistics of national accounts as comparative data in order to eliminate breaks in time-series data. Timmer et al. (2013) explained WIOT’s country and industry classification and the process of deriving each country’s IO tables.

## 3. Research methodology

### 3.1 Research model

At the center of the 4th Industrial Revolution changes is “manufacturing innovation,” which constructs systems for digitalization, servicitization of products and customized production through an automatic and smart production process using ICT technology. The convergence of the machinery·equipment industry and the ICT industry is a prerequisite for the implementation of smart factories. Thus, it is necessary to compare the inter-industry linkage effect of the ICT and machinery·equipment industries in the USA, Germany, Japan, China and Korea, the five countries that are promoting national innovative strategies to prepare for the 4th Industry Revolution and to provide political implications by examining the global competitiveness of each country and the spillover effect on all industries. Figure 1 presents the research model of this study.

### 3.2 Hypothesis

In 2014, the US domestic market size of the ICT industry was the largest in the world, which was valued at $1.11 trillion and took up 30 percent of the world’s total. China’s ICT domestic market has grown with the focus of device and communication services, which is reaching$377.9bn and accounts for 8.5 percent. Japan accounted for 8 percent and ranked third, following by Germany, which accounted for 3.9 percent and ranked fourth. The market size of Korea was $77.5bn, accounted for 2.1 percent of the world’s total and ranked 9th in the world (Gartner, 2014). China was the largest ICT exporter of the world in 2013; the amount of exports was$508.4bn accounted for 31.8 percent of global exports. The USA ranked the second place with the amount of $140bn and accounted for 8.7 percent. Korea ranked the fourth place with$107.1bn and accounted for 6.7 percent. Germany ranked the seventh place, and Japan ranked the eighth, both accounted for 3.9 percent.

Since ICT industries in Korea, China, the USA, Germany and Japan have different domestic market sizes, export scales, growth rates, and manufacturing and service productions, their influence on different domestic industries is expected to differ. Particularly, the innovation process of the ICT ecosystem is operated better in the USA than in EU and other developing countries (Fransman, 2009), and the accelerated productivity in the USA has been based on ICT service instead of ICT manufacturing since the mid-1990s (Basu and Fernald, 2006). Thus, the USA is expected to have high forward linkage effect but low backward linkage effect. In Korea, ICT industry captured 8.7 percent of the GDP in 2013 and took up 23 percent of the real growth rate of GDP in the second quarter of 2014, which is expected to have a considerably high inter-industry linkage effect. Besides, since ICT manufacturing has shown its strong points, its backward linkage effect is expected to be higher than that in other countries. Because ICT manufacturing in China is strong, its backward linkage effect is expected to be higher than other countries. In contrast, both the backward and forward linkage effects of ICT is expected to have low in Germany, since the proportion of ICT manufacturing decreased, and ICT service sector does not stand out. We propose the following hypotheses:

H1.

Inter-industry linkage effects of ICT industry are different in Korea, China, the USA, Germany and Japan.

H1a.

Backward linkage effects of ICT industry are different in Korea, China, the USA, Germany and Japan.

H1b.

Forward linkage effects of ICT industry are different in Korea, China, the USA, Germany and Japan.

UN Comtrade showed Germany, the USA, Japan and China at the first, second, third and fourth places, respectively, in the world in exports of general machinery. Korea ranked eighth by accounting for 3.4 percent of the world’s total. In terms of imports, the USA ranked first, China second, Germany third, Korea eighth and Japan tenth worldwide. Annual average increase in the demand for general machinery in the global market was 9.8 percent between 2001 and 2011. During the same period, the annual average increase in export amount in China, Korea, the USA, Japan and Germany was 10.5 percent. China recorded rapid growth in export with an annual average growth rate of 27.8 percent, followed by Korea with an annual average growth rate of 17.7 percent. According to OECD, machinery·equipment investment in Japan was very high, whereas that in the USA was not high compared to its economic scale. Machinery is the production-based sector of manufacturing industry and the key of manufacturing facility. It has a high backward linkage effect since it is a B2B industry that manufactures various materials and components.

The inter-industry linkage effects of the machinery·equipment industry are different in Korea, China, the USA, Germany and Japan, due to the distinctions between each country’s proportion of the machinery·equipment industry, exports, industrial structure, etc. Moreover, backward linkage effects in China and Korea is expected to be even higher than those in the other countries since its manufacturing share is exceedingly high. In contrast, manufacturing in the USA has little importance, so the forward and backward linkage effects of the machinery·equipment industry are expected to be lower than those in other countries. We propose the following hypotheses:

H2.

Inter-industry linkage effects of the machinery·equipment industry are different in Korea, China, the USA, Germany and Japan.

H2a.

Backward linkage effects of the machinery·equipment industry are different in Korea, China, the USA, Germany and Japan.

H2b.

Forward linkage effects of the machinery·equipment industry are different in Korea, China, the USA, Germany and Japan.

The proportion and spillover effect of the ICT industry and the machinery·equipment industry differ among Korea, China, the USA, Germany and Japan since each country has different industrial structures. In general, the ICT industry has higher forward linkage effect than backward linkage effect, and thus provides many supports and supplies to the overall production activities. Therefore, forward linkage effect is expected to be higher than backward linkage effect in all of Korea, China, the USA, Germany and Japan. Since the machinery·equipment industry is the key industry that provides the equipment base to manufacturing industry (Arnold, 2001). In other words, because the machinery·equipment industry uses various materials and components to boost production activities, Korea, China, the USA, Germany and Japan are all expected to have high backward linkage effect.

In general, the ICT industry has high forward linkage effect, while machinery·equipment industry has high backward linkage effect. Especially, since the USA has a strong ICT service, which helps boost the productivity of other industries, the difference between the forward linkage effects of ICT and that of the machinery·equipment industry is expected to be greater than that in other countries. We propose the following hypotheses:

H3.

The inter-industry linkage effects between the ICt and machinery·equipment industries are different in Korea, China, the USA, Germany and Japan.

H3a.

The backward linkage effects between the ICT and machinery·equipment industries are different in Korea, China, the USA, Germany and Japan.

H3b.

The forward linkage effects between the ICT and machinery·equipment industries are different in Korea, China, the USA, Germany and Japan.

### 3.3 Research process

The research process involves: collecting the amount of intermediate consumption of each industry from WIOT of Korea, China, the USA, Germany and Japan released by WIOD; classifying the machinery·equipment industry and re-classifying the ICT industry by adding all ICT-related industries into one; using Excel to compute backward and forward linkage effects of the ICT industry and the machinery·equipment industry; using SPSS 18.0 to conduct one-way ANOVA to verify each hypothesis; and analyzing the difference between the two industries and the five countries (Figure 2).

#### 3.3.1 Data collection

This study uses WIOT that covers 56 industries in 43 countries. Although the amount of inputs in both the ICT industry and machinery equipment industry decreased in the USA, Korea, Germany and Japan in 2009 due to the global financial crisis in 2008, the ICT and machinery equipment industries of all countries except Germany’s ICT industry returned to recovery in 2010. China reduced the amount of inputs in ICT industry but then recovered in 2010, while the machinery equipment industry did not decrease even in 2009. Although each country has different industrial structures and growth patterns that may affect the results, there was no distortion of the data that have been used to compare between the countries because all the countries were affected by the financial crisis in 2008. Industries C26 (manufacture of computer, electronic, and optical products), J61 (telecommunications) and J62_63 (computer programming, consultancy and related activities; information service activities) on WIOT are classified as the ICT industry according to OECD’s ICT industry classification. On the other hand, industry C28 (manufacture of machinery and equipment n.e.c.) on WIOT is classified as the machinery·equipment industry according to the international standard International Standard Industrial Classification (ISIC) Rev.4 (UN, 2006).

#### 3.3.2 Industry classification

The data of WIOT use each country’s released statistics, considering the ease of verification and possibility of composing additional statistics in the future. The supply-use tables of each country were standardized to 59 products and 35 industries. Each of the product and industry classifications were determined based on Classification of Products by Activity and Nomenclature statistique des activités économiques dans la Communauté européenne, the classification system established in EU. UN’s ISIC is composed by UNSD and is used as the industrial standard in most countries in the world. Rev.4 published in 2006 classified industries into four broad and detailed structures (UN, 2006).

In 2007, OECD’s Working Party on Indicators for the Information Society improved the classification standard of ICT industry in ISIC Rev.4.0 by dividing it into manufacturing and service sectors, which enabled cross-national comparison on economic activities of ICT industry (OECD, 2006, 2009). Both ISIC Rev.4.0 and WIOT-classified machinery·equipment industry is the same as Division 28 (manufacture of machinery and equipment n.e.c.).

#### 3.3.3 Analysis method

Inter-industry analysis, which is also called IO analysis, uses IO tables to quantitatively examine the inter-relationship between industries. IO tables describe the overall economic activities within an economy for a certain period (generally one year) using matrices to clearly organize and record how the products and services of a particular industry are distributed to other industries or sectors and how outputs of other industries or sectors are inputted to each industry for production based on certain rules, so that people can view the inter-industry relationships within an economy at a glance (Kin, 2015). IO analysis uses Leontief inverse matrix coefficients and equations to derive each industry’s inducement coefficient that induces the demand of raw materials and intermediary goods to examine each industry’s spillover effect. The production inducement coefficient refers to the total units of outputs required, both directly and indirectly, when the final demand of production is increased by one unit. The inter-industry linkage effect refers to the result of dividing each industry’s production inducement effect by the average inducement effect of all industries. Such inter-industry linkage effect, or spillover effect, can be divided into backward and forward linkage effects (Hirschman, 1958). Although there are various methodologies for measuring backward and forward linkage effects, this study adopted the commonly used Rasmussen method (Rasmussen, 1957).

Changes in production activities in one industry create demand for input in related productions, and productions of the related industries will be influenced simultaneously due to such indirect inducement. The total of direct and indirect inducement effect is represented by demand balanced formula X=(IA)−1d. That is, the IO matrix (marked as A) is derived by dividing each column in the IO table by the total output of the corresponding industry. In order to find out the demand balanced condition of each production when the final demand is determined externally, the vector of production (X), vector of final demand (d), and the identity matrix I are used together with A to form the formula (IA)x=d. Thus, if (IA) is a non-singular matrix, it will have an inverse matrix. The formula is X=(IA)−1d and is called the Leontief inverse matrix. The result represents the coefficient of inducement effect of final demand and composition change on the production activities of each industry. Therefore, the formula to compute backward linkage effect is B L j = 1 / n i B i j / 1 / n 2 i j B i j , where i B i j is the sum of the column elements of the Leontief inverse matrix (IA)−1. Similarly, the forward linkages can be obtained from the rows of the Leontief inverse matrix, the formula being F L i = 1 / n i B i j / 1 / n 2 i j B i j , where j B i j is the sum of the row elements of the Leontief inverse matrix.

## 4. Results

### 4.1 Spillover effect of ICT

This study used one-way ANOVA to test all hypotheses and ran Levene’s test to verify the assumption of homogeneity of variance. Levene test showed that the value of ICT backward linkage effect is 3.468 and the significance probability is 0.012, whereas the result of Forward linkage effect is 5.135 and the significance probability is 0.001. The results indicate that the null hypothesis is rejected and the homogeneous variance assumption in invalid, thus Tamhane’s T2 was conducted after Welch’s ANOVA test (Figure 3 and Table III).

The difference between Korea and China, the USA and Germany is not statistically significant, with an F-value of 51.015 and p<0.001, whereas one-way ANOVA found significant differences between all other countries, in the order of Korea≑China>Japan>the USA≑Germany. In the USA, backward linkage effect continued to decrease from 1.0652 in 2001 to 0.8893 in 2014. Korea has the highest backward linkage effect, which increased from 1.1028 in 2000 to 1.1494 in 2009, but then started decreasing and eventually dropped to 1.0546 in 2014. Backward linkage effect in China exceeded Korea since 2010, and it was the highest among all the five countries in 2014 with the value of 1.1307. ICT industry has a high forward linkage effect since it continuously creates new demands in the market. Likewise, the results in this study showed that the forward linkage effects of the ICT industry in Korea, China, the USA, Germany and Japan were all higher than 1 in the order of Japan ≑> the USA≑Korea ≑> China ≑> Germany. Germany’s ICT industry was found to have lower forward linkage effect than Japan, the USA and Korea. Forward linkage effect in the USA kept reducing from 2.2297 in 2000 to 1.7953 in 2014, while that in Japan remained stably high from 2.0471 in 2000 to 1.9337 in 2014. In comparison with forward linkage effect in the USA, the forward linkage effect in Japan was higher and more stable.

### 4.2 Spillover effect of the machinery·equipment industry

Levene test showed that backward linkage effect in the machinery·equipment industry is 2.076 and the significance probability is 0.093, meaning that the homogeneous variance assumption is valid. Thus, Scheffe’s test was conducted. On the other hand, forward linkage effect is 25.156 and the significance probability is 0, indicating that the homogeneous variance assumption is invalid, and hence Tamhane’s T2 was conducted after Welch’s ANOVA test (Figure 4 and Table IV).

Backward linkage effects of the machinery·equipment industry are high in all five countries in the order of China>Japan≑Korea>the USA>Germany. However, backward linkage effect in Germany decreased year after year from 1.558 in 2001 to 1.0124 in 2014, which was lower than that observed in the remaining four countries. On the other hand, although backward linkage effect in the USA was lower than that in Korea and Japan, it increased slightly from 1.1176 in 2000 to 1.1279 in 2014.

Differences in forward linkage effects among the five countries are statistically significant, in the order of China>Korea>Germany≑Japan≑the USA Unlike the ICT industry, forward linkage effects of the machinery·equipment industry are mostly low (less than 1), while China and Korea is the country that has high forward linkage effects (greater than 1). The USA has the lowest forward linkage effect among the five countries. Germany and Japan had similar forward linkage effect until 2009, but since then, Germany has started to have a higher forward linkage effect than Japan year after year. Although forward linkage effect in China was extremely high, it was gradually decreasing. On the other hand, forward linkage effect in Korea showed upward trend.

See Table V.

## 5. Discussion and conclusion

### 5.1 Conclusions

See Table VI.

The forward linkage effect of the ICT industry is higher than the backward linkage effect in all countries, indicating ICT’s high innovation effect on the supply side to boost the growth of other industries. The reason is that rapid technology innovation does not only happen within the ICT industry, but also stimulates other industries to accelerate emergence and servicitization between industries and foster the development of new industries. Demand goes on in other industries such as the servicitization of manufacturing industries, and the trend is expected to continue in the future.

Machinery·equipment industry has higher backward linkage effect than forward in four countries. However, forward linkage effect in China is even higher, while both backward and forward linkage effect exceed 1, implying high inducement effect. Furthermore, since both backward and forward linkage effect in the machinery equipment industry of China are overwhelmingly higher than other countries, continuous investment on achinery equipment industry will be a good investment direction for invigorating all industries in China.

### 5.2 Implications and limitations

The IO analysis will play a big role in determining the important or leading industries in the national economy based on each industry’s spillover effect, as well as in providing significant standards for analysis to determine the priority of investing using limited resources. As presented in Figure 5, the inducement effects in the ICT and machinery·equipment industries can be divided into two groups clearly. Since both backward and forward linkage effects of the ICT and machinery·equipment industries in China and Korea exceed 1, active government investments are strongly recommended because it will help invigorate all industry through its inducement effect on other industries. Especially, the intentional convergence of the ICT and machinery·equipment industries will increase mutual synergy, and thus help China to become the “manufacturing powerhouse” that it is pursuing, and it will enable response to the 4th Industrial Revolution through the Smart Factory projects in Korea.

In the USA, the forward linkage effect of ICT service industry is high, whereas the backward linkage effect of the ICT industry and the spillover effect of machinery·equipment industry are relatively low. Thus, the current innovation strategy in manufacturing industry to prepare for the 4th Industrial Revolution appears suitable, which is a strategy based on a global platform that utilizes the strong cloud computing power. The spillover effect of ICT industry is the lowest in Germany among the five countries. Despite Germany having the greatest manufacturing capabilities in the world, its enterprises may remain as subcontractors of US platform companies if the convergence with the domestic ICT industry is delayed. Therefore, Germany must increase the pace of construction of effective and smooth smart production systems by strengthening ICT services, and seek strategies to increase the inter-industry spillover effect. The ICT service industry and the machinery·equipment industry in Japan are in a stable state. Japan has already had a good foundation for constructing the high-tech economic and social system that merges the ICT industry with robotics. In contrast to Germany, which pursues innovation in the entire manufacturing industry, Japan is concentrated in the robotics sector. Thus, the scope of innovation may be limited, although the synergy effect is expected to be high.

This study identifies the industrial structure that establishes a foundation of the Industrial Revolution, and makes academic and practical contributions by trying for the first time to derive each country’s strengths and weaknesses to provide implications on policy formulation through quantitative comparison with developed countries. The IO analysis provides the standard for analyzing a key industry based on the spillover effect of each industry. However, the production inducement effect computed using IO table failed to consider direct effects such as R&D investment, which may lead to underestimation. On the other hand, the effect may be overestimated since it is a static analysis that does not consider changes in price. Therefore, since it is difficult to make conclusive evaluations on relative size, people should be cautious toward the practical applications of the results and additional research is needed to overcome this limitation.

In addition, it is difficult to understand all infrastructure industries for the 4th Industrial Revolution in each country by only examining the spillover effects of the ICT and machinery·equipment industries; thus, future studies may include biomedical industries and divide the ICT industry into manufacturing and service to conduct a more detailed research. Furthermore, to examine the argument that the backward linkage effects of ICT in Japan and Korea are decreasing due to the overseas expansion of the two countries’ ICT component enterprises, it is necessary to conduct further analysis that considers imports. Finally, it will also be meaningful for future studies to examine the economic influence of the convergence between the machinery·equipment and ICT industries in each country.

## Figures

Research model

Research process

#### Figure 3

Comparison of backward and forward linkage effects of ICT industry

#### Figure 4

Comparison of backward and forward linkage effects of machinery·equipment

#### Figure 5

Industrial spillover effects of the ICT industry and machinery·equipment industry in Korea, China, the USA, Germany and Japan

## Table I

Existing research on the 4th Industrial Revolution

Type Researcher Contents
General Porter and Heppelmann (2014) The USA is leading and obtaining benefits from smart connectivity
Drath and Horch (2014) Addressed an easy approach to the core idea of Industry 4.0 and industrial requirements for success
Application Alemanno (2017) The problems of information control and imbalance of use due to data retention
Weiss et al. (2016) Showed the limitations on US ability of three case studies through experiments on the cooperation between humans and robots
Education (Suganya, 2017) Global higher education changes for the 4th Industrial Revolution
Smithers and James (2017) Three pillars for effective communication according to technological change of the 4th Industrial Revolution: English for Business Purposes (EBP), Computer-mediated Communication (CMC), English as a Business Lingua Franca (BELF)
Employment Arntz et al. (2017) Future labor research on the 4th Industrial Revolution, which examined the influence of automation technology on the total labor market
Graetz and Michaels (2015) Although the growth rate of industrial robotics increased in 17 countries, it did not have new impact on total employment
Manufacturing innovation Mosterman and Zander (2016) Address concrete examples of CPS
Lee et al. (2015) Proposal of an integrated five-level architecture as a guideline for implementing CPS: smart connection – data to information conversion – cyber – cognition – configuration
Monostori (2014) CPS may lead to the 4th Industrial Revolution

## Table II

Research on the ICT and machinery·equipment industries

Industry Type Researcher Contents
ICT industry General Yang et al. (2013) The entry barriers of ICT firms are lower than those of non-ICT firms, and the manufacturing sector finds a stronger innovation effect than the service sector
Garcia and Vincente (2014) Analyzed the capabilities of ICT in Europe using the structure hole theory
Country Jung et al. (2013) Verified the hypothesis that technology convergence is the main motivation of the recent productivity increment in Korea, and estimated the effect of ICT on total factor productivity (TFP)
Xing et al. (2011) Analysis of convergence status and roles of ICT manufacture and ICT service industries of China for the period of 2002
Fransman (2009) Showed that the USA operated the innovation process of ICT ecosystem better than Europe and other developing countries
Kim (2014) Strategies for improving global competitiveness and strengthening overseas expansion of the ICT industry through fundamental changes and innovations for revitalize the Japanese economy
Machinery·equipment industry General Arnold (2001) Took a general view of the history of machine tool industry, and explained and noted the industry’s sharp changes caused by the adoption of numerical control system
Richter and streb (2011) German machine tool traders blamed China on patent infringement, while Germany itself used imitation strategy until the 1920s
Country Shinno et al. (2006) A pair-wise comparison and quantitative SWOT analysis of the global competitiveness of Japan’s machine tool industry using a matrix
Kalafsky and Macpherson (2002) Most American manufacturers in the machine tool industry are small businesses, but small-sized enterprises can hardly flourish in the long term
Ernst (1995) Systematically evaluated 50 businesses within Germany’s mechanical engineering industry
Fransman (1986) Addressed how the machine tool fields in Taiwan and Japan were affected by changes in technology and production, the mechanism of growth and the role of the nation

## Table III

Comparison of backward and forward linkage effects of ICT industry

ICT KOR CHN USA GER JPN ICT KOR CHN USA GER JPN
2000 1.1028 1.0923 1.0318 0.9133 1.0644 2000 1.7941 1.6587 2.2297 1.4996 2.0471
2001 1.0940 1.0852 1.0652 0.9140 1.0631 2001 1.8921 1.7556 2.1663 1.5683 2.0327
2002 1.1229 1.0559 1.0434 0.9058 1.0416 2002 1.9615 1.7496 2.0806 1.6016 2.0222
2003 1.1164 1.0257 1.0062 0.9594 1.0334 2003 1.8996 1.7248 1.9918 1.6570 2.0353
2004 1.1319 1.0416 0.9741 0.9377 1.0259 2004 1.9162 1.8757 1.8973 1.6469 2.0036
2005 1.1308 1.0854 0.9458 0.9563 1.0156 2005 1.9244 2.0548 1.8456 1.6716 1.9223
2006 1.1465 1.0552 0.9392 0.9353 1.0037 2006 2.0012 1.8627 1.7882 1.6433 1.9140
2007 1.1352 0.9972 0.9174 0.9428 0.9954 2007 1.9534 1.5567 1.7783 1.6825 1.9105
2008 1.1183 1.0328 0.8991 0.9530 0.9887 2008 1.7931 1.4499 1.7720 1.6998 1.9825
2009 1.1494 1.0729 0.9192 0.9758 0.9550 2009 1.9420 1.5019 1.8013 1.7748 1.9353
2010 1.0878 1.1193 0.9146 0.9542 0.9676 2010 1.7425 1.7603 1.8204 1.6871 1.9819
2011 1.0588 1.1464 0.9209 0.9299 0.9740 2011 1.5150 1.8005 1.8434 1.6641 1.9935
2012 1.0682 1.1426 0.9225 0.9038 0.9760 2012 1.6624 1.8204 1.8620 1.6114 1.9938
2013 1.0593 1.1378 0.8855 0.8912 0.9684 2013 1.6641 1.8391 1.7553 1.6116 1.9888
2014 1.0546 1.1397 0.8893 0.8925 0.9685 2014 1.6621 1.8250 1.7953 1.5940 1.9337
ICT in five countries F=51.015*** ICT in five countries F=17.241***
Korea≑China>Japan>the USA≑Germany Japan ≑> the USA≑Korea ≑> China ≑> Germanya

Notes: a≑>means A country≑B country or A country>B country. *p<0.05; **p<0.01; ***p<0.001

## Table IV

Comparison of backward and forward linkage effects of the machinery·equipment industry

Machinery KOR CHN USA GER JPN Machinery KOR CHN USA GER JPN
2000 1.1584 1.2816 1.1176 1.0552 1.2015 2000 0.9662 1.6452 0.8608 0.9183 0.9108
2001 1.1783 1.2826 1.1124 1.0558 1.2193 2001 0.9355 1.6757 0.8522 0.9212 0.9069
2002 1.1839 1.2745 1.1312 1.0459 1.2387 2002 0.9372 1.6777 0.8576 0.9064 0.8801
2003 1.1782 1.2716 1.1377 1.0555 1.2099 2003 0.9726 1.6472 0.8527 0.9334 0.8797
2004 1.1843 1.2798 1.1204 1.0550 1.2077 2004 0.9835 1.6165 0.8362 0.9218 0.9106
2005 1.2147 1.2857 1.1182 1.0532 1.1934 2005 1.0231 1.4876 0.8380 0.9136 0.9282
2006 1.2266 1.2994 1.1226 1.0523 1.1900 2006 1.0972 1.5637 0.8352 0.9212 0.9208
2007 1.2215 1.3085 1.1208 1.0355 1.1911 2007 1.1595 1.6066 0.8246 0.9045 0.9291
2008 1.1773 1.3160 1.1258 1.0408 1.1801 2008 1.1075 1.5665 0.8222 0.9080 0.9116
2009 1.1672 1.3183 1.1347 1.0367 1.1631 2009 1.0725 1.6021 0.8526 0.8784 0.8516
2010 1.1845 1.3095 1.1400 1.0066 1.1593 2010 1.2060 1.4595 0.8261 0.8734 0.8536
2011 1.1770 1.3116 1.1377 1.0145 1.1769 2011 1.2164 1.4638 0.8359 0.8906 0.8751
2012 1.1562 1.3199 1.1569 1.0182 1.1795 2012 1.1449 1.3322 0.8566 0.8998 0.8893
2013 1.1512 1.3334 1.1340 1.0127 1.1586 2013 1.1295 1.3783 0.8619 0.9013 0.8821
2014 1.1609 1.3408 1.1279 1.0124 1.1429 2014 1.1391 1.3537 0.8622 0.9085 0.8531
Machinery in five countries F=324.063*** Machinery in five countries F=251.544***
China>Japan≑Korea>the USA>Germany China>Korea>Germany≑Japan≑the USA

Notes: *p<0.05; **p<0.01; ***p<0.001

## Table V

Comparison of backward and forward linkage effects between the ICT industry and the machinery·equipment industry

## Table VI

Summary of results

Hypothesis Result Contents
A1 (ICT) A1a Accepted Backward linkage effect: Korea≑China>Japan>the USA≑Germany
Backward linkage effect in Korea was the highest, but has been on the downtrend since 2009
Coefficient of the USA reduced from 1.0652 in 2001 to 0.8893 in 2014
A1b Accepted Forward linkage effect: Japan ≑> the USA≑Korea ≑> China ≑> Germany Forward linkage effect in each country are all higher than 1
Forward linkage effect in Japan remained high, while that in the USA continued to decrease
Backward linkage effect of China exceeded that in Korea since 2010
A2 (machinery·equipment) A2a Accepted Backward linkage effect: China>Japan≑Korea>the USA>Germany
Backward linkage effect of China maintained that highest since 2000
A2b Accepted Forward linkage effect: China>Korea>Germany≑Japan≑the USA
Forward linkage effect of China is the highest, and forward linkage effect of China and Korea are higher than 1
A3 (ICT/machinery·equipment) A3a Accepted ICT in all five countries: backward linkage effect<forward linkage effect

## References

Alemanno, A. (2017), “Consumers or citizens? How the 4th Industrial Revolution can help people change law and policy”, Discussion Paper, Digital Society-Education, Inclusion, and Jobs, Friends of Europe, Brussels, pp. 71-75.

Arnold, H. (2001), “The recent history of the machine tool industry and the effects of technological change”, Institute for Innovation Research and Technology Management, University of Munich, Munich, 2001-2014.

Arntz, M., Gregory, T., Lehmer, F., Matthes, B. and Zierahn, U. (2017), “Technology and jobs in the fourth industrial revolution”, JEL, J21, J23, D24, O33.

Basu, S. and Fernald, J. (2006), “Information and communications technology as a general-purpose technology: evidence from U.S industry data”, Working Paper No. 2006-29, San Francisco, CA.

Drath, R. and Horch, A. (2014), “Industry 4.0: hit or hype?”, Industry Forum: IEEE Industrial Electronics Magazine, Vol. 8 No. 2, pp. 56-58.

Ernst, H. (1995), “Patenting strategies in the German mechanical engineering industry and their relationship to company performance”, Technovation, Vol. 15 No. 4, pp. 225-240.

Fransman, M. (1986), “International competitiveness, technical change and the state: the machine tool industry in Taiwan and Japan”, World Development, Vol. 14 No. 12, pp. 1375-1396.

Fransman, M. (2009), The New ICT Ecosystem: Implications for Policy and Regulation, Cambridge University Press, New York, NY.

Garcia, M.A.S. and Vincente, M.R. (2014), “ICT technologies in Europe: a study of technological diffusion and economic growth under network theory”, Telecommunication Policy, Vol. 38 No. 4, pp. 360-370.

Gartner (2014), “Gartner market databook”, 4Q14 Update, available at: www.gartner.com/doc/2953117?ref=mrktg-srch (accessed November 15, 2017).

Graetz, G. and Michaels, G. (2015), “Robots at work”, CEP Discussion Paper No. 1335, Centre for Economic performance, pp. 2042-2695.

Hirschman, A.O. (1958), The Strategy of Economic Development, Yale University Press, New Haven, CT, 58.

Jung, H.J., Youn, K.Y. and Yoon, C.H. (2013), “The role of ICT in Korea’s economic growth: productivity changes across Industry since the 1990s”, Telecommunications Policy, Vol. 37 No. 4, pp. 292-310.

Kalafsky, R.V. and Macpherson, A.D. (2002), “The competitive characteristics of U.S manufacturers in the machine tool industry”, Small Business Economics, Vol. 19 No. 4, pp. 355-369.

Kim, T.-E. (2014), “Japan’s ICT international competitiveness strengthening international development initiative report”, Institute of Information and Telecommunication Policy, Vol. 26 No. 16, pp. 20-34.

Kin, T.H. (2015), 2010 and 2013 Regional Inter-Industry Input Table, Korea Bank, Seoul.

Kwak, K.H. and Park, J.Y. (2009), “Analysis of the Korean machinery industry using Korea’s input-output analysis”, Journal of Industrial Economics and Business, Vol. 22 No. 1, pp. 179-199.

Lee, J., Bagheri, B. and Kao, H.-A. (2015), “A cyber-physical systems architecture for Industry 4.0-based manufacturing systems”, Manufacturing Letters, Vol. 3, pp. 18-23.

Lee, J.K. (2017), “Characteristics and implications of ICT industry in Korea”, Hyundai Research Institute, Vol. 675 No. 2, pp. 1-11.

Mattioli, E. and Lamonica, G.R. (2013), “The ICT role in the world economy: an input-output analysis”, Journal of World Economic Research, Vol. 2 No. 2, pp. 20-25.

Monostori, L. (2014), “Cyber-physical production systems: roots, expectations and R&D challenges”, Procedia CIRP, Vol. 17, pp. 9-13.

Mosterman, P.J. and Zander, J. (2016), “Industry 4.0 as a cyber-physical system study”, Software and Systems Modeling, Vol. 15 No. 1, pp. 17-29.

OECD (2006), ICT Sector Definition, Presentation for 10th Meeting of the WPIIS OECD, DSTI/ICCP/IIS(2006)/FINAL, Paris.

OECD (2009), Guide to Measuring the Information Society, OECD, Paris.

OECD (2017), The Next Production Revolution: Implications for Governments and Business, OECD Publishing, Paris, (accessed November 12, 2017).

Porter, M.E. and Heppelmann, J.E. (2014), “How smart, connected products are transforming competition”, Harvard Business Review, HBR Reprint R1411C, November, pp. 1-23.

Rasmussen, P.N. (1957), “Studies in inter-sectoral relations”, The American Economic Review, Vol. 47 No. 3, pp. 432-435.

Richter, R. and Streb, J. (2011), “Catching-up and falling behind: knowledge spillover from American to German machine toolmakers”, The Journal of Economic History, Vol. 71 No. 4, pp. 1006-1031.

Schwab, K. (2016), “The fourth industrial revolution: what it means, how to respond”, Davos 2016, available at: www.weforum.org/agenda/2016/01/the-fourth-industrial-revolution-what-it-means-and-how-to-respond/ (accessed October 20, 2017).

Shinno, H., Hachiag, S., Yoshioka, H. and Marpaung, S. (2006), “Quantitative SWOT analysis on global competitiveness of machine tool industry”, Journal of Engineering Design, Vol. 177 No. 3, pp. 251-258.

Song, J.Y. and Kwak, K.H. (2012), “Machine industry ICT convergence and value added effect”, Journal of Mechanical Science and Technology, Vol. 52 No. 11, pp. 37-42.

Statistics Korea (2017), “Korea’s machinery equipment industry”, available at: http://kosis.kr/statisticsList/statisticsListIndex (accessed November 20, 2017).

Suganya, G. (2017), “A study on challenges before higher education in the emerging fourth industrial revolution”, International Journal of Engineering Technology Science and Research, IJETSR, Vol. 4 No. 10, pp. 1-3.

UN (2006), “International Standard of Industrial classification of All Economic Activities(ISIC): Revision 4”, Statistical Papers Series M No. 4, Economic & Social Affairs.

Weiss, A., Huber, A., Minichberger, J. and lkeda, M. (2016), “First application of robot teaching in an existing Industry 4.0 environment: does it really work?”, Societies, Vol. 6 No. 3, pp. 1-21.

Xing, W., Ye, X. and Kui, L. (2011), “Measuring convergence of China’s ICT industry: an input-output analysis”, Telecommunications Policy, Vol. 35 No. 4, pp. 301-303.

Yang, C., Lee, S.-G. and Lee, J. (2013), “Entry barrier’s difference between ICT and non-ICT industries”, Industiral Management & Data Systems, Vol. 113 No. 3, pp. 461-480.

Alberto, B.-M., Margarita, B. and Fernado, L.-L. (2013), “Perceived performance effects of ICT in manufacturing SMEs”, Industrial Management & Data Systems, Vol. 113 No. 1, pp. 117-135.

Dietzenbacher, E., Stehrer, R., Timmer, M. and Vries, G. (2012), “Trade performance in internationally fragmented production networks: concepts and measures”, WIOD Working Paper No. 11, Groningen.

Korea Bank (2015), Inter-Industry Analysis Commentary, Korea Bank, Seoul.

Timmer, M.P., Los, B. and Vries, G.J. (2016), “An anatomy of the global trade slowdown based on the WIOD 2016 release”, Groningen Growth and Development Centre, University of Groningen, Groningen, No. GD-162.

Timmer, M.P., Temursho, U., Streicher, G., Stehrer, R., Rueda, J.M., Pindyuk, O., Los, B., Francois, J.F., Vries, G.J. and Arto, I. (2012), “The World Input-Output database (WIOD): contents, sources and methods”, WIOD Working Paper No. 10, WIOD, Groningen.

Ya-Ching, L., Pin-Yu, C. and Hsien-Lee, T. (2011), “Corporate performance of ICT-enabled business process re-engineering”, Industrial Management & Data Systems, Vol. 111 No. 5, pp. 735-754.

## Corresponding author

Sang-Gun Lee can be contacted at: slee1028@sogang.ac.kr