Macroeconomic Determinants of India’s Participation in Global Value Chains: An Empirical Evidence

Abstract

Global value chains (GVCs) have significantly changed the world trade scenario. Many developed countries gained benefits through production fragmentation, and it has worked as hope for the developing world. India is also one of the participants but its share in global GVC space is very limited. The recent pandemic comes with a lot of opportunities for India to emerge as a new GVC hub in the Asia-Pacific region. This requires the study of factors determining the extent of India’s participation in GVCs. The present article is an attempt in this direction wherein various factors determining the forward and backward participation levels are identified. At a macro level, the study found the positive role of technology advancement, domestic capital and industrial capacity in promoting the level of participation in GVCs, while the role of net FDI inflows is found to be negative. This highlights the requirement for conducive policies to reap the maximum benefits associated with foreign capital. India is already observing the benefits and growing, but still, it has to cover a long path. The time has come to make an image of the brand India in the world and reap maximum benefits from the GVC participation.

JEL Codes: F14, F40, O24

Keywords

Global value chains India forward participation backward participation ARDL

Introduction

The nature of trade saw significant changes due to the rise of production fragmentation, especially since the last decade of the twentieth century. Major factors that compelled the fragmentation involved reduction in trade costs, technological advancements, trade and economic liberalisation (Amador & Cabral, 2016). This gave rise to the phenomenon of Global value chains (GVCs). It involves the series of stages involved in the production of the commodity from its conception to its final form provided multiple countries are involved in the production process. At present, more than 60% of global trade consists of the trade of intermediate inputs (Ramaswamy et al., 2021). Goods are produced at different locations adding value at each stage at each location.

The rise of GVCs offered new opportunities for both developed and developing countries. The developed world outsources the tasks to the developing world which involves the usage of low-cost inputs. On the other hand, for developing countries, GVCs worked as an opportunity to amalgamate the global production process without developing a full manufacturing base. This in return also provided the developed countries to focus more on the research and development aspect of product development (Gereffi et al., 2005) and help developing countries generate more economic opportunities including income and employment. UNCTAD (2013) emphasises the significance of participation in GVCs for developing countries. It outlines the benefits associated with developing countries in terms of increased opportunities for employment, increase in productivity of local enterprises and exports. Further, it reduces the burden on developing countries to develop a full industrial base rather than focus on the process in which it enjoys a comparative advantage. Several developing countries transformed their economy by participating in GVCs. For instance, China by assembling manufactured parts and components, Vietnam through the electronics industry, Bangladesh through textile industries and Taiwan through the electronic industry.

India being a developing country often considered a country that has ‘missed the boat’. As per the study by Athukorala (2019), India failed to capture the slice of the production process that arises due to global production fragmentation. China captured the gain by becoming a global hub for multi-national enterprises (MNEs) to outsource the assembling component of their production. This might be due to their low labour cost and the creation of Special Economic zones (SEZs). However, India too had created SEZ and availability of cost-effective labour. But its structural composition remains different. Unlike China, in India, SEZ benefits were only confined to exchange control and tax rebate. No attempt was made to tune in domestic rules, regulations and procedures in synchronisation with MNE cultures and their best practices. Looking at its labour abundance, it was expected that India could do better in terms of labour-intensive manufacturing. However, it did well in a few capital-intensive manufacturing and skill-intensive services sectors. As per the OECD (2014), India’s participation in regional value chains, with the neighbouring countries, is the lowest in the developing world. This highlights the need to improve the participation level to reap maximum benefits from this economic activity.

GVCs are a more granular form of trade measurement than traditional trade. It decomposes gross exports or traditional exports into domestic and foreign value added (Antras, 2016). In the empirical literature, it is often computed as a summation of the backward and forward participation levels of the country. Backward participation involves the share of foreign inputs in the gross output or exports of the country and forward participation includes the share of domestic value added in the gross output or exports of the other country. Backward participation/integration outlines the demand side of the economy while forward participation denotes the economic strength of the country. From the global trend, it is seen that the level of participation of each country is different depending upon many factors. The basic research question associated with the rise of GVCs has often sought attention ‘What explains the rise of GVCs?’. According to the World Development Report 2020, improvement in information and communication technology (ICT), reduction in trade and non-trade barriers, and realisation of many trade agreements have provided a conducive environment for growth in GVCs. However, when it comes to the country level, the factors determining the level and type of participation differ.

Given this belief, the present study highlights the India-specific factors affecting its participation in GVCs using time-series data from the year 1990 to 2019. Due to the paucity of data points, the study considers only the main macroeconomic variables that are affecting the level of participation as per the conceptual framework given in the next section. The study attempted to determine the factors affecting both backward and forward participation in GVCs. To pursue the study’s objectives, the entire study is organised into five sections including the present introductory one. Section 2 provides the conceptual framework highlighting various determinants of GVC participation with support from the literature. Section 3 provides the sources of data, the construction of variables for the empirical analysis, and the econometric method used for the empirical analysis. Section 4 presents the empirical findings and discusses the results. The last section concludes.

Factors Determining GVCs Participation: A Conceptual Framework

Traditionally, to the classical, trade has been viewed as a source to overcome the comparative disadvantage faced by the country. Varied resource availability in different regions/countries was the main basis of trade among countries. There exist various explanations for trade in finished products from different industries. Differences in cost advantages are the main reason behind trade under perfect competition. Over time, globalisation and liberalisation with the advancement of technology make intra-industry trade possible. Paul Krugman in his new trade theory explains the reasons behind intra-industry trade and emphasises the role of scale economies under the existence of monopolistic competition. Technology made it possible to produce parts and components in different countries and globalisation made it possible to trade these parts and components for making the final product. Under globalisation, countries also thought more trade towards more prosperity. Following the literature, many factors are responsible for the rise in trade in parts and components, sometimes it is known as trade in intermediate goods which gave rise to GVCs. Those factors include globalisation, trade liberalisation, technological advancement, differences in endowments, a conducive policy environment and distance to the GVC hub, among others.

The main factor responsible for the rise in GVCs is globalisation and trade liberalisation. As mentioned earlier, globalisation connects countries. It is globalisation with the help of which various trade integration efforts have been made by almost all countries of the world. These efforts reduce the trade and investment barriers among countries and increase competition which induces producers to look outside the country to reduce costs (Deardorff, 1998). Reduction in trade costs proved to be one of the major reasons for the expansion of GVCs. According to the literature, trade cost impacts get magnified along value chains. Major channels of reduction in trade costs involve the signing of trade agreements, reduction in tariff and non-tariff barriers, and inclination of the country towards trade openness policies. Since GVC itself is the most granular form of trade, therefore, trade cost reduction promotes GVCs integration (Baldwin, 2011a, 2011b).

The impact of falling trade costs is further enhanced through technological development. The ICT revolution proved to be the stepping stone in the expansion of GVCs in the 1990s. However, the recent composition of technology provided greater concern for GVCs integration. It is being pointed out that due to advancements in technology including 3D printing, the internet of things, big data, blockchain technologies, and robotics may perturb the labour cost advantage enjoyed by the developing world. This increasing automation may compel the developed world to re-offshore their production process from the developing world. Moreover, it might also lead to greater demand for skilled workers and thereby exacerbating the divide between North and South which could have been bridged through GVC integration. However, another strand of literature advocates the positive role of technology in GVC integration. For instance, increased internet penetration in Africa and China leads to a dramatic rise in export and employment levels (Hjort & Poulsen, 2019). Further, the development of new logistics systems, such as blockchain technology and the internet of things, helps to improve the delivery system by monitoring shipments of products on a real-time basis. According to MGI (2019), advancements in logistics technologies could reduce customs processing and shipment time by 16%–28%.

Further, differences in factor endowments are another main factor responsible to enhance GVC trade. Natural endowment impact positively the forward participation of the country. Conceptually speaking, countries with higher natural resources often produce goods that lie under upstream industries. These goods are exported to other countries which process them before culminating in final goods, increasing the forward participation of the country. However, in the case of backward participation effect seems to be negative. This is because resource-intensive industries do not require many imports of the intermediate from other countries. Several studies substantiate the above relation including (Romalis, 2004), (Fernandes et al., 2020) and (Chakrabarty & Chanda, 2021). The availability of domestic capital further intensifies cross-border production fragmentation. A study by Adarov and Stehrer (2021) empirically verifies that capital accumulation enhances backward participation in most sectors. Further, investment in ICT capital generally enhances the backward integration of technology-intensive sectors including electrical, chemical and transport sectors. Similarly, a study by Pathikonda and Farole (2017) argues that investment in long-term capabilities has the potential to shape GVC participation in the country. Long-term capabilities include human and capital investment. This can enhance GVC integration, especially in technology-intensive sectors. Conceptually speaking, domestic capital helps in industrialising the country. This helps in developing an appetite for the country to import basic and intermediate goods and add value to them subsequently, and as a result enhancing its backward integration. In the case of forward participation, results seem to be inconclusive (Adarov & Stehrer, 2021). This might be because forward participation involves the exports of goods which are used as an intermediate in the other countries. So, if country occupies ‘central position’ in the GVCs production stage, then it might produce mostly downstream products. As a result, its forward integration gets reduced. However, there can be other way round also. If country produces goods that lies relatively on the upstream stages of production, then it can enhance its forward integration.

In addition, foreign capital such as foreign direct investment (FDI) also shapes the GVC participation of the country. Theoretically, inward flows of FDI work as a catalyst especially for developing countries to produce in concordance with MNEs and increase their GVC integration. Studies including (Buelens & Tirpák, 2017), (Pathikonda & Farole, 2017) and (Amendolagine et al., 2017) examine the impact of FDI on GVCs participation. Overall, these studies emphasise that FDI can enhance the technology frontier of the economy, spread foreign knowledge and reduce capital scarcity prevalent in the host country. This also brings through MNEs intermediate goods required to add value to the goods to be produced in the FDI-inflowing countries. As a result, it increases backward integration in FDI-inflowing countries. However, in the case of forward integration, FDI may impact negatively too. This might be because of the failure of MNEs to invest in the value addition of the host country’s exports (Kowalski et al., 2015).

Both availabilities of natural resources and capital determine the industrial capacity of a country which plays an important role in shaping the GVCs. The industrial capacity of the economy holds a positive association with forward participation. This can be understood by the logic that as a country industrialises, it generates the capacity to produce varieties of goods in its domestic economy, and, therefore, enhancing its exporting appetite. However, backward participation is impacted negatively by industrial capacity. The development of the industrial base reduces the import dependence from other countries. However, countries with large industrial bases often capture the ‘central position’ in the GVCs’ stages of production. These countries often specialise in downstream industries and import basic and intermediate goods from the rest of the world. As a result, increasing their backward integration (Antràs & Gortari, 2020). Hence, it works in both ways depending upon the production activities of a country.

In addition to these common factors, there exist many other factors that affect countries differently. These include geopolitical risks, a distance of a country from a major GVC hub in the region, the availability of advanced trade infrastructure and the business environment, among others. Overall, all the major factors affecting a country’s participation in GVCs are highlighted in Figure 1.

Figure 1.

Source: Authors’ own construction.

For the Indian economy, there exist few studies that have studied various factors determining participation in GVCs. A recent study by Ray and Miglani (2020) based on the survey data also highlighted that India has low integration into GVCs in the manufacturing sector although it can integrate more. The paper also highlights the main issue of the limited role played by the lead firm that governs the entire value chain and sells the final product. Aggarwal et al. (2021) while studying the sectoral TiVA database provided by the OECD, highlighted the role of domestic capital, foreign capital and labour intensity in promoting more participation in GVCs. These studies are focused on sectoral aspects. However, the literature becomes scant when it comes to macroeconomic determinants of India’s participation in GVCs. The present study tried to fill this gap and adds to the literature by studying the differential factors determining forward and backward participation separately using the autoregressive distributed lag (ARDL) framework with time-series data.

Construction of Variables and Methodology

Construction of Variables

The major issue with the empirical analysis is the measurement of the GVC participation rate. The measure should not be plagued by the problem of double counting. To cater to this problem, several databases provide the participation rate such as EORA, OECD-TiVA, ADB-MRIO and WIOD. However, our study chose the EORA database because of its greater and the latest data availability. EORA provides two measures of GVC participation: DVX and FVA from 1990 to the year 2018. Where DVX is the measure of forward participation, and FVA is the measure of backward participation. Two separate models have been estimated, one for each level of participation, to determine the macroeconomic determinants of India’s participation in GVCs.

Further, the selection of independent variables draws out from the literature reviewed in the previous section and is also based on regular data availability. The study has taken five independent variables as per the conceptual framework. All the variables are taken from the World Development Indicators database provided by the World Bank. A list of explanatory variables with their measurement is given in Table 1.

Table 1.

Construction of Variables.

Variables (Notation)	Measure	Represent
Gross capital formation (GCF)	Gross fixed capital formation plus the change in stocks	Domestic capital availability/formation
FDI	Net FDI inflows	Inflows of foreign capital
Natural resources (Res)	Natural resource rent is the sum of oil rents, natural gas rents, coal rents (hard and soft), mineral rents and forest rents	Availability of natural resources
Patents (Pat)	Number of patents registered by residents and non-residents in a country	Proxy variable to represent the technological advancement
Industrial capacity (IndCap)	Industrial value added at constant prices	Capacity to manufacture goods

Source: Authors’ construction.

All variables are taken at constant prices for the year 2015. In addition to this, for the time-series analysis, a log of all the variables has been taken for ease of interpretation and to achieve stationarity at a lower level of integration.

Methodology

As per the data availability and objective of the study, the present study has used the ARDL model developed by Pesaran et al. (2001) for checking the relationship among time-series variables. The method has the utility of greater efficiency in case of a small sample size, and when the variables are of a different order of integration. To pursue the study’s objectives, two main models have been estimated—one with forward participation as the dependent variable and the other with backward participation as the dependent variable. The specification of estimated model-1 in a general ARDL specification is given as follows:

Δ {L D V X}_{t} = ϑ_{0} + δ_{1} {L D V X}_{t - 1} + δ_{2} {L G C F}_{t - 1} + δ_{3} {L F D I}_{t - 1} + δ_{4} {L R e s}_{t - 1} + δ_{5} {L P a t}_{t - 1} + δ_{6} {L I n d C a p}_{t - 1} + \sum_{i = 1}^{p} σ_{i} Δ {L D V X}_{t - i} + \sum_{i = 1}^{q} α_{i} {Δ L G C F}_{t - i} + \sum_{i = 1}^{r} β_{i} {Δ L F D I}_{t - i} + \sum_{i = 1}^{s} {_{i}}_{} Δ {L R e s}_{t - i} + \sum_{i = 1}^{n} μ_{i} {Δ L P a t}_{t - i} + \sum_{i = 1}^{m} ρ_{i} Δ {L I n d C a p}_{t - i} + ε_{t}

(1)

where L represents the log of the variables; $σ, α, β,, μ, ρ$ are the short-run coefficients; $δ^{'} s$ are the long-run coefficients; $ϑ$ is the constant; p, q, r, s, n and m are optimal lag orders for each variable.

Similarly, the specification of model-2 estimated for backward participation is given as follows:

Δ {L F V A}_{t} = ϑ_{0} + δ_{1} {L F V A}_{t - 1} + δ_{2} {L G C F}_{t - 1} + δ_{3} {L F D I}_{t - 1} + δ_{4} {L R e s}_{t - 1} + δ_{5} {L P a t}_{t - 1} + δ_{6} {L I n d C a p}_{t - 1} + \sum_{i = 1}^{p} σ_{i} Δ {L F V A}_{t - i} + \sum_{i = 1}^{q} α_{i} {Δ L G C F}_{t - i} + \sum_{i = 1}^{r} β_{i} {Δ L F D I}_{t - i} + \sum_{i = 1}^{s} {_{i}}_{} Δ {L R e s}_{t - i} + \sum_{i = 1}^{n} μ_{i} {Δ L P a t}_{t - i} + \sum_{i = 1}^{m} ρ_{i} Δ {L I n d C a p}_{t - i} + ε_{t}

(2)

The bounds test checks the existence of co-integration by checking the joint significance of long-run coefficients in the above equations. In case of the presence of co-integration, the following error correction model provides the short-run dynamic coefficients of the model’s adjustment to long-run equilibrium:

Δ {L D V X}_{t} = ϑ_{0} + \sum_{i = 1}^{p} σ_{i} Δ {L D V X}_{t - i} + \sum_{i = 1}^{q} α_{i} {Δ L G C F}_{t - i} + \sum_{i = 1}^{r} β_{i} {Δ L F D I}_{t - i} + \sum_{i = 1}^{s} {_{i}}_{} Δ {L R e s}_{t - i} + \sum_{i = 1}^{n} μ_{i} {Δ L P a t}_{t - i} + \sum_{i = 1}^{m} ρ_{i} Δ {L I n d C a p}_{t - i} + ω {E C T}_{t - 1} + ε_{t}

(3)

where $ω$ is the speed of adjustment parameter with a value less than one, and a negative sign implies the convergence towards a long-run equilibrium relationship. ECT_t-1 is an error correction term constructed from the residuals of the long-run equation. However, in case of no co-integration, then the following model is used to estimate the short-run coefficients.

Δ {L D V X}_{t} = ϑ_{0} + \sum_{i = 1}^{p} σ_{i} Δ {L D V X}_{t - i} + \sum_{i = 1}^{q} α_{i} {Δ L G C F}_{t - i} + \sum_{i = 1}^{r} β_{i} {Δ L F D I}_{t - i} + \sum_{i = 1}^{s} {_{i}}_{} Δ {L R e s}_{t - i} + \sum_{i = 1}^{n} μ_{i} {Δ L P a t}_{t - i} + \sum_{i = 1}^{m} ρ_{i} Δ {L I n d C a p}_{t - i} + ε_{t}

(4)

The same specification is used to estimate the same for backward participation. In the end, both the models have been tested for stability and other diagnostics. The empirical results obtained from the estimation are given in the next section.

Empirical Findings

The empirical findings in the study are presented in two subsections. The first section shows the descriptive analysis of the level of India’s participation in GVCs, its sectoral contribution and major GVC partners. The second section presents the empirical results of the ARDL estimation determining factors affecting forward and backward participation in GVCs for the Indian economy.

India and the GVCs

Level of India’s GVC Participation

Figure 2 reports the trend of forward and backward participation from the year 2000. It is seen from the figure that the level of backward participation remains in the range of 10%–15% and the level of forward participation between 25% and 30%. Overall, the level of GVC participation in India remains between 38% and 48%.

Figure 2.

Source: UNCTAD-EORA GVC database.

Sectoral Composition of India’s Participation in GVCs

Further, Table 2 shows the sector-wise bifurcation of forward participation to get a clear picture. In India, major sectors contributing to the GVC exports by forward participation are renting related to business activities (18.54 %) followed by coke and petroleum (15.60 %), chemicals (12.49 %), metals (7.56 %), textiles products (6.08 %) and transport equipment (5.02 %) among other sectors. These sectors together contribute 65% of India’s forward GVCs exports.

Table 2.

Composition of India’s GVC Exports by Forward Participation: Sector-wise (2018).

Sector	Description	Share (%)
1	Renting of M&Eq and other business activities	18.54
2	Coke, refined petroleum and nuclear fuel	15.60
3	Chemicals and chemical products	12.49
4	Basic metals and fabricated metal	7.56
5	Textiles and textile products	6.08
6	Transport equipment	5.02
7	Mining and quarrying	4.09
8	Retail trade, except motor vehicles and motorcycles, repair of household goods	3.57
9	Agriculture, hunting, forestry and fishing	3.22
10	Electrical and optical equipment	3.14
11	Rubber and plastics	2.64
12	Manufacturing, nec and recycling	2.63
13	Machinery and nec	2.39
14	Water transport	1.63
15	Inland transport	1.50
16	Food, beverages and tobacco	1.45
17	Post and telecommunications	1.34
18	Wholesale trade and commission trade, except for motor vehicles and motorcycles	1.27
19	Air transport	1.22
20	Other supporting and auxiliary transport activities, activities of travel agencies	1.08
21	Leather, leather products and footwear	0.91
22	Other nonmetallic minerals	0.78
23	Wood and products of wood and cork	0.50
24	Pulp, paper, paper products, printing and publishing	0.47
25	Financial intermediation	0.33
26	Sale, maintenance and repair of motor vehicles and motorcycles, retail sale of fuel	0.33
27	Construction	0.13
28	Other community, social and personal services	0.12
29	Hotels and restaurants	0.01
30	Electricity, gas and water supply	0.01
31	Education	0.00
32	Health and social work	0.00
33	Public administration and defence, compulsory social security	0.00
34	Real estate activities	0.00
35	Private households with employed persons	0.00

Source: Authors’ calculations based on Multi-Region Input–Output Database, ADB.

Moreover, Table 3 shows major sectors contributing to the GVC exports by backward participation are coke and petroleum (27.05%) followed by metals (13.07%), transport equipment (12.24%), chemical products (8.49) and textile products (5.55 %) among other sectors. The top three sectors constitute 51% of India’s total GVC exports by backward participation.

Table 3.

Composition of India’s GVC Exports by Backward Participation: Sector-wise (2018).

Sector	Description	Share
1	Coke, refined petroleum and nuclear fuel	27.05
2	Basic metals and fabricated metal	13.07
3	Transport equipment	12.24
4	Chemicals and chemical products	8.49
5	Textiles and textile products	5.55
6	Manufacturing, nec and recycling	5.51
7	Renting of M&Eq and other business activities	5.31
8	Inland transport	4.08
9	Electrical and optical equipment	3.58
10	Machinery and nec	3.56
11	Food, beverages and tobacco	2.47
12	Rubber and plastics	1.92
13	Mining and quarrying	1.39
14	Other nonmetallic minerals	1.13
15	Leather, leather products and footwear	0.89
16	Agriculture, hunting, forestry and fishing	0.53
17	Retail trade, except motor vehicles and motorcycles, repair of household goods	0.51
18	Other community, social and personal services	0.39
19	Pulp, paper, paper products, printing and publishing	0.37
20	Air transport	0.36
21	Water transport	0.32
22	Wholesale trade and commission trade, except for motor vehicles and motorcycles	0.31
23	Other supporting and auxiliary transport activities, activities of travel agencies	0.27
24	Wood and products of wood and cork	0.17
25	Construction	0.16
26	Post and telecommunications	0.14
27	Hotels and restaurants	0.09
28	Electricity, gas and water supply	0.05
29	Financial intermediation	0.05
30	Sale, maintenance and repair of motor vehicles and motorcycles, retail sale of fuel	0.04
31	Health and social work	0.00
32	Education	0.00
33	Real estate activities	0.00
34	Public administration and defence, compulsory social security	0.00
35	Private households with employed persons

Source: Authors’ calculations based on Multi-Region Input–Output Database, ADB.

The sectoral analysis reveals that in a few sectors India’s participation in GVCs is strong. The strong manufacturing sectors that have the potential to participate further are chemical and chemical products, business services and transport equipment. Many other sectors in the top 10 categories also have the potential to do better in terms of increased participation if sufficient attention has been given to the issues and challenges faced by those sectors. The main sectors under this category could be transport equipment sectors (including all kinds of transport), electrical equipment and machinery, among others. In almost all sectors, India is facing competition from other countries in supplying intermediates as well as final products to other countries. These industries are also facing many domestic issues that are hindering their further growth.

Major Partners

Although India’s contribution to GVCs trade remains minuscule as compared to the global trade related to GVCs. Its total GVC-related trade is around $140 billion, out of which nearly 10% is directed towards Singapore (SGP), followed by the United States (USA 6.9%), China (PRC 3.52%) and Germany (GER 3.50%), among others as shown in Figure 3. The figure depicts the major GVC partners of India in terms of the gross exports decomposed into forward and backward GVC participation. Values obtained are computed for GVC participation in bilateral exports between India and the rest of the major GVC partner countries (for more details see Belotti et al., 2020).

Figure 3.

Source: Authors’ construction from ADB-MRIO data for the year 2018 using STATA.

Factors Determining India’s Participation in GVCs

To determine factors affecting India’s participation in GVCs, the present study has estimated two models—one for forward participation and the other for backward participation. The results of the empirical analysis are given in the following subsections:

Checking for Stationarity

Given the time-series nature of the data, the estimation starts with checking of stationarity of each variable. This is done through various unit-root tests. Checking stationarity requires the determination of the optimal lag length of each variable. The same is done for each variable, and the optimal lag length to be found is one for all variables except the industrial capacity whose optimal lag length is two. The lag length is determined based on the Akaike information criterion and the Schwarz criterion values. Following this, unit-root tests are used to check the order of integration of each series at which the series becomes stationary if it is not at levels. The present study has used ADF and DF-GLS unit-root tests for this purpose, and the results are shown in Table 4.

Table 4.

Test Statistic for Checking Stationarity of the Variables.

Variable	At Levels		At First Difference
Variable	ADF	DF-GLS	ADF	DF-GLS
LDVX	−1.606	−1.029	−3.420	−3.927
LFVA	−1.728	−1.113	−3.434	−3.898
LGCF	−0.963	−1.544	−3.171	−2.305
LFDI	−4.245	−3.776	–	–
LRes	−1.003	−1.681	−4.155	−4.148
LPat	−1.238	−2.255	−4.219	−4.223
LIndCap	−1.087	−2.598	−3.083	−4.848

Source: Authors’ calculations in STATA 16.

Notes: Results of unit-root tests are reported for the model with no drift and no time trend; The results for other models are not reported because of space and could be available at the reader’s request; Critical values at the 1% and 5% levels of significance for ADF test are by MacKinnon are −3.736 and −2.994, respectively; Critical values at the 1% and 5% level of significance for DF-GLS test are −3.770 and −3.360, respectively.

The analysis of unit-root tests indicates that all the variables are integrated of order one, that is, I(1) except FDI which is stationary at level. This result supports the application of the ARDL method to our data where one can use a mix of I(0) and I(1) variables together to find the short-run and long-run relationship among the variables under study.

Descriptive Statistics

Table 5 provides the summary statistics of all the variables used in the econometric analysis. The logarithm form of variables has been used to report the summary statistics.

Table 5.

Summary Statistics.

Variable	Observations	Mean	Standard  Deviation	Minimum  Value	Maximum Value
LDVX	30	17.087	1.11	15.226	18.418
LFVA	30	16.265	1.265	13.998	17.747
LGCF	30	26.396	0.766	25.184	27.493
LFDI	30	22.903	1.597	18.668	24.585
LRes	30	24.119	0.681	23.145	25.258
LPat	30	9.726	0.971	8.139	10.890
LIndCap	30	26.391	0.570	25.531	27.261

Source: Authors’ calculations in STATA 16.

Further, the stationary variables are used to construct the correlation matrix of all the variables considered for the analysis purpose. Based on the unit-root test results in Table 4, all the variables are taken at first difference except FDI inflows. The correlation matrix is shown in Table 6.

Table 6.

Correlation Matrix.

Variables	$Δ L D V X$	$Δ L F V A$	$Δ L G C F$	$L F D I$	$Δ L R e s$	$Δ L P a t$	$Δ L I n d C a p$
$Δ L D V X$	1.000
$Δ L F V A$	0.895	1.000
$Δ L G C F$	0.236	0.091	1.000
$L F D I$	−0.195	−0.307	0.173	1.000
$Δ L R e s$	0.786	0.768	0.057	0.069	1.000
$Δ L P a t$	0.355	0.331	0.027	0.030	0.367	1.000
$Δ L I n d C a p$	0.308	0.173	0.578	0.191	0.128	0.289	1.000

Source: Authors’ calculations in STATA 16.

It reveals from the correlation matrix that forward and backward participation is highly correlated with each other, followed by the availability of natural resources in the country. With other variables, both types of GVC participation are positively correlated except for FDI inflows in which case GVC participation and FDI inflows are negatively correlated. The results reveal that in the case of India, net FDI inflows are not helping to increase the level of GVC participation.

Testing Co-integration Using Bound Test

Further, to test the existence of long-run relationships among variables the study has used the Pesaran et al. (2001) bounds test. The test statistics assume no level relationship as a null hypothesis means null of no co-integration with an alternate of the existence of co-integration. Table 7 presents the value of the F-statistic based on which the existence of co-integration is decided. The results reveal that the F-value for both models is not significant to reject the null hypothesis of no co-integration. Hence, the bounds test confirms the non-existence of long-run relationships among the variables under study. The reason behind the inconclusiveness of this result is the annual data with few observations. The test may find the long-run relationship in the monthly or quarterly data from the study period but given the annual frequency of data and that too for 30 years, test statistics failed to conclude.

Table 7.

ARDL Bounds Test.

Panel I: Bound Testing to Co-integration: Model 1 Forward Participation  Estimated Model: LDVX = f (LGCF, LFDI, LRes, LPat, LIndCap)
Indicator	Value
F-statistic	3.387
Panel II: Bound Testing to Co-integration: Model 2 Backward Participation  Estimated Model: LFVA = f (LGCF, LFDI, LRes, LPat, LIndCap)
F – Statistic	4.570

Source: Authors’ calculations.

Notes: Kripfganz and Schneider’s (2018) critical values at the 5% and 1% levels of significance at I(1) are 5.039 and 7.412, respectively.

Estimation of Short-run Relationship: ARDL Model Estimation

After confirming the non-existence of the long-run relationship, the short-run estimates are calculated using the ARDL specification given in equation (4), and the results are given in Tables 8 and 9. Table 8 shows the ARDL model results for forward participation.

Table 8.

Factors Determining Forward Participation.

Dependent Variable: LDVX	Coefficient	P Value
Short-run Coefficients—ARDL (1,1,1,1,1,2)
${Δ L D V X}_{t - 1}$	0.527**	0.014
$Δ L G C F$	0.177	0.509
$Δ L F D I$	−0.104*	0.073
$Δ L R e s$	0.333***	0.001
$Δ L P a t$	0.140	0.150
$Δ L I n d C a p$	0.926	0.221
Constant	−3.121	0.616
Diagnostic Tests
F-statistics	756.30	0.000
BG LM test (Chi²)	1.011	0.315
White test (ARCH) (Chi²)	26.46	0.151

Source: Authors’ calculations.

Notes: *, ** and*** denote the significance of coefficient at the 10%, 5% and 1%, respectively; the Null hypothesis of the BG LM test is no serial correlation; the Null hypothesis of the White test is no heteroscedasticity.

The estimation results reveal that the past year’s value of forward participation is positively and significantly impacting the current year’s level of forward participation. This shows that over time, India’s participation in forward GVCs is rising. Further, the significant coefficient of natural resources shows that natural endowment/resources in India play a significant role in determining the extent of forward GVC participation. Results are in concordance with the cross-country determinants of forward GVC participation. It corroborates with the findings of (Fernandes et al., 2020). In other words, a country (here, India) with a higher endowment of natural resources leads to higher forward GVC participation. This justifies the theoretical framework that natural-resource-intensive commodities are processed further in downstream industries, thereby, adding indirect values to Indian exports through other countries. However, FDI inflows in the case of India are negatively impacting the extent of forward participation. Theoretically speaking, FDI bridges the technological, physical capital and human capital gap through its spillover effects. But over here, negative impact may implicate that FDI inflows are not intended to spur the value addition of the host country’s exports (Kowalski et al., 2015). In the case of India, this may be due to the market-seeking nature of FDI which are less into exports and more into selling in the host country. All other variables are positively impacting forward participation, but the impact is not statistically significant. Further, diagnostic tests have been performed to check the robustness of the results. The study has reported F-statistic for the overall significance of the model, Durbin Watson d-test and Breusch-Godfrey LM test for autocorrelation and White test for the presence of heteroscedasticity. The diagnostic results reveal that the estimated model is free from the problem of heteroscedasticity and autocorrelation.

On the other hand, Table 9 provides the estimates of determinants of backward GVC participation. The results highlight that over the years on average the backward participation of India in GVCs is rising as shown by the positive and significant lagged coefficient of foreign value added. Further, the negative and significant coefficient of FDI inflows reveals the negative impact of FDI inflows on backward participation. In the case of India, this may be due to the excessive restrictions on the usage of the host country’s raw material/intermediate products and reinvestment of earned profits in the host country.

Table 9.

Factors Determining Backward Participation.

Dependent Variable: LFVA	Coefficient	P Value
Short-run Coefficients—ARDL (1,1,1,1,1,2)
${Δ L F V A}_{t - 1}$	0.590***	0.005
$Δ L G C F$	−0.169	0.598
$Δ L F D I$	−0.189**	0.012
$Δ L R e s$	0.504***	0.000
$Δ L P a t$	0.225*	0.065
$Δ L I n d C a p$	1.774*	0.056
Constant	−4.363	0.496
Diagnostic Tests
F-statistics	639.72	0.000
BG LM test (Chi²)	3.195	0.0739
White test (ARCH) (Chi²)	26.71	0.1437

Source: Authors’ calculations.

The positive and significant coefficient of natural resource availability in India is also not as per the economic theory. It may be because the sector using natural resources contributes almost 16% to total forward participation and almost 27% to backward participation and is ranked number one among all the sectors (see Tables 2 and 3). For this model, assuming a 10% level of significance, the number of patents and industrial capacity is positively impacting the extent of backward participation revealing that over time India is progressing towards various medium and heavy industries whose components are imported from other countries. Sectoral analysis in Table 3 supports the fact that the transport sector is at the third position from the top and constitutes almost 12% of the total backward participation of India. This also supports the positive impact of technology on backward participation in the Indian economy.

Causality Analysis

Further, to verify the causality among the variables, the VAR model at the first difference has been estimated and causality has been tested using the Granger causality test statistic. The results for both models are reported in Table 10. Panel A in Table 10 shows the existence of causal relationships among the variables included in model 1 where forward participation and other variables have been considered for the final estimation, and panel B shows the existence of causality among the variables included in model 2, where factors determining backward participation are considered. Panel C shows the causality between forward and backward GVC participation.

Table 10.

Granger Causality Test Results.

Panel A			Panel B			Panel C
Model 1 with Forward Participation			Model 2 with Backward Participation			Forward and Backward Participation
Equation	Excluded	P Value	Equation	Excluded	P Value	Equation	Excluded	P Value
$Δ L D V X$	$Δ L G C F$	0.037**	$Δ L F V A$	$Δ L G C F$	0.019**	$Δ L F V A$	$Δ L D V X$	0.064*
	$L F D I$	0.009***		$L F D I$	0.018**
	$Δ L I n d C a p$	0.025**		All	0.000***
	All	0.000***	$Δ L R e s$	$Δ L G C F$	0.017**
$Δ L R e s$	$Δ L G C F$	0.028**		$Δ L I n d C a p$	0.049**
	All	0.000***		All	0.000***
$Δ L P a t$	$Δ L D V X$	0.015**	$Δ L P a t$	$L F D I$	0.011**
	$L F D I$	0.001***		$Δ L R e s$	0.040**
	$Δ L R e s$	0.003***		$Δ L I n d C a p$	0.027**
	$Δ L P a t$	0.005***		All	0.004***
	All	0.000***

Source: Authors’ calculations.

Note: Results are shown only for those variables in between cause and effect relationship exist statistically; The detailed results could be available at the reader’s request; *, ** and *** denote the significance of the coefficient at the 10%, 5% and 1%, respectively.

The results in panel A reveal that the lagged values of domestic capital availability, inflows of foreign capital and capacity to manufacture goods cause forward participation. However, the lagged values of forward participation cause only technological advancement. In addition, the lagged values of domestic capital availability also cause the availability of natural resources. Further, the results in panel B show that the lagged values of domestic capital availability and inflows of foreign capital caused backward participation whereas the lagged values of backward participation are not causing to any variable. In this case, too, the lagged values of domestic capital availability and industrial capacity also causing to natural resources. In addition, the technological advancement represented by patents is caused by lagged values of foreign capital inflows, availability of natural resources and industrial capacity. Finally, the results given in panel C show that it is the lagged values of forward participation which is causing backward participation but not vice versa.

Model Stability

Finally, to check the stability of the estimated models, the study has used CUSUM’s square test whose output is given in Figure 4. As shown, the CUSUM statistic is within the 5% critical bounds, which implies that the estimated coefficients obtained from the ARDL model are stable.

Figure 4.

Source: Authors’ construction using STATA 16.

Conclusions

The last decade of the twentieth century saw stupendous changes in trading relations among the countries. GVCs have significantly changed the scenario of world trade. The main reasons for its growth are better market access across borders and varying costs of inputs over countries. Better market access facilitates the growth of GVCs through trade in intermediate goods (or parts and components). Declining trade costs over time have also worked as a positive factor behind the growth of these value chains globally. From the trade point of view, GVCs saw a significant rise due to rapid globalisation. There are many other factors responsible for the rise of GVCs. The question of identifying country-specific factors determining the level of participation induces us to pursue the present study.

With the help of secondary data on domestic value added, foreign value added and economy-wide variables impacting the extent of GVC participation, the present study found the positive role of technology advancement, domestic capital and industrial capacity in promoting the level of participation in GVCs while the role of net FDI inflows is found to be negative. This highlights the requirement for conducive policies to reap the maximum benefits associated with foreign capital. India is already observing the benefits and growing along the path but still, it has to cover a long path. The time is come to make an image of the brand India in the world and reap maximum benefits from the GVC participation. The study has taken limited factors because of two reasons: a small number of observations and a lack of regular data over the study period. The study results could be improved by finding out the determinants using the panel data over many countries for the study period. Also, the India-specific analysis could be improved by analysing sector-specific factors.

Footnotes

Acknowledgement

The authors are grateful for the anonymous referee’s comments. Views are the authors own.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

References

Adarov

, & Stehrer

(2021). Implications of foreign direct investment, capital formation and its structure for global value chains. The World Economy , 44(11), 3246–3299.

Aggarwal

, Chakraborty

, & Bhattacharyya

(2021). Determinants of domestic value added in exports: Empirical evidence from India’s manufacturing sectors . SAGE Publications.

Amador

, & Cabral

(2016). Global value chains: A survey of drivers and measures. Journal of Economic Surveys , 30(2), 278–301. https://doi.org/10.1111/joes.12097

Amendolagine

, Presbitero

, Rabellotti

, Sanfilippo

, & Seric

(2017). FDI, global value chains, and local sourcing in developing countries. IMF Working Papers , 17(284), 1–40. https://doi.org/10.5089/9781484335109.001

Antras

(2016). Global production: Firms, contracts, and trade structure . Princeton University Press. http://press.princeton.edu/titles/10613.html

Antràs

, & Gortari

(2020). On the geography of global value chains. Econometrica , 88, 1553–1598. https://doi.org/10.3982/ECTA15362

Athukorala

(2019). Joining global production networks: Experience and prospects of India. Asian Economic Policy Review , 14(1), 123–143.

Baldwin

(2011a). 21st century regionalism: Filling the gap between 21st century trade and 20th century trade rules (Issue ERSD-2011-08). https://econpapers.repec.org/RePEc:zbw:wtowps:ersd201108

Baldwin

(2011b). Trade and industrialisation after globalisation’s 2nd unbundling: How building and joining a supply chain are different and why it matters . https://doi.org/10.3386/w17716

10.

Belotti

, Borin

, & Mancini

(2020). icio: Economic analysis with inter-country input–output tables in Stata. World Bank Policy Research Working Paper, 9156.

11.

Buelens

, & Tirpák

(2017). Reading the footprints: How foreign investors shape countries’ participation in global value chains. Comparative Economic Studies , 59(4), 561–584. https://doi.org/10.1057/s41294-017-0036-2

12.

Chakrabarty

, & Chanda

(2021). What drives global value chain participation? SSRN Electronic Journal , Working Paper, No. 650, 1–47. https://doi.org/10.2139/ssrn.3946183

13.

Deardorff

A. V.

(1998). Fragmentation in simple trade models. Discussion Paper, No. 422, Research Seminar in International Economics.

14.

Fernandes

, Kee

H. L.

, & Winkler

(2020). Determinants of global value chain participation: Cross-country evidence.. https://doi.org/10.1596/1813-9450-9197

15.

Gereffi

, Humphrey

, & Sturgeon

(2005). The governance of global value chains. Review of International Political Economy , 12(1), 78–104.

16.

Hjort

, & Poulsen

(2019). The arrival of fast internet and employment in Africa. American Economic Review , 109(3), 1032–1079.

17.

Kowalski

, Lopez-gonzalez

, Ragoussis

, & Ugarte

(2015). Participation of developing countries in global value chains : Implications for OECD trade policy papers No. 179 participation of developing countries in global value chains. January 2016. https://doi.org/10.1787/5js33lfw0xxn-en

18.

Kripfganz

, & Schneider

D. C.

(2018). Estimating autoregressive distributed lag and  equilibrium correction models . In Proceedings of the 2018 London Stata Conference, 59.

19.

OECD. (2014). Enhancing global value chains participation, India Policy Brief, OECD Better Policies Series.

20.

Pathikonda

, & Farole

(2017). The capabilities driving participation in global value chains. Journal of International Commerce, Economics and Policy , 08, 1750006. https://doi.org/10.1142/S1793993317500065

21.

Pesaran

M. H.

, Shin

, & Smith

R. J.

(2001). Bounds testing approaches to the analysis of level relationships. Journal of Applied Econometrics , 16(3), 289–326.

22.

Ramaswamy

, Choutagunta

, & Sahu

S. K.

(2021). Evaluating Asian free trade agreements: What does gravity model tell us? Foreign Trade Review , 56(1), 60–70. https://doi.org/10.1177/0015732520961330

23.

Ray

, & Miglani

(2020). India’s GVC integration: An analysis of upgrading efforts and facilitation of lead firms. Working Paper, No. 386, Indian Council for Research on International Economic Relations (ICRIER), New Delhi, India.

24.

Romalis

(2004). Factor proportions and the structure of commodity trade. American Economic Review , 94(1), 67–97. https://doi.org/10.1257/000282804322970715

25.

UNCTAD. (2013). Global value chains and development: Investment and value-added trade in global economy, United Nations.