ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

FDI Spillovers and Export Performance of Indian Manufacturing Firms after Liberalisation

The spillovers from foreign direct investment through multinational enterprises have attracted considerable attention in recent times. Existing empirical studies on fdi spillovers largely look at the productivity enhancing effects and horizontal spillovers of foreign firms in the same industry sector ignoring the possibility of spillovers through buyer-supplier or backward linkages. The present study examines the impact of horizontal as well as backward spillovers from the presence of foreign firms, on the export performance of domestic firms in the Indian manufacturing industry during 1993-2008. Increased competition in the domestic market post-liberalisation through sales of foreign firms is forcing domestic firms to look for export markets. The results indicate that domestic firms are not benefited in improving their export performance through any buyer-supplier linkages with the mnes.

SPECIAL ARTICLE

FDI Spillovers and Export Performance of Indian Manufacturing Firms after Liberalisation

T J Joseph, V Nagi Reddy

The spillovers from foreign direct investment through multinational enterprises have attracted considerable attention in recent times. Existing empirical studies on FDI spillovers largely look at the productivity enhancing effects and horizontal spillovers of foreign firms in the same industry sector ignoring the possibility of spillovers through buyer-supplier or backward linkages. The present study examines the impact of horizontal as well as backward spillovers from the presence of foreign firms, on the export performance of domestic firms in the Indian manufacturing industry during 1993-2008. Increased competition in the domestic market post-liberalisation through sales of foreign firms is forcing domestic firms to look for export markets. The results indicate that domestic firms are not benefited in improving their export performance through any buyer-supplier linkages with the MNES.

T J Joseph (tjjoseph@gmail.com) is at the INC Research Staff College, Hyderabad and V Nagi Reddy (vnreddy1941@gmail.com) is at the ICFAI Institute for Management Teachers, Hyderabad.

Introduction

T
he extent of globalisation of a country’s economy is usually evaluated from its trade and investment relations with the rest of the world, especially the volume and growth of trade from that country. Exports help firms to achieve greater e fficiency in production through economies of scale due to increased market size. Studies point out that exports may also improve the innovative activities of the firm, with new varieties of products and new methods of delivery to be competitive in quality and to stay in the business (Aggarwal 2002; Kumar and Pradhan 2003). Sjoholm (1999) points out that for an exporter to be successful in a foreign market requires good knowledge about the foreign market conditions such as foreigners’ preferences, regulations, distribution channels and other market characteristics. However, collecting information on some of the above-mentioned variables may usually be costly and this may deter entry of firms into the foreign market. Sjoholm indicates that there can be substantial reduction in sunk entry cost with various types of foreign contacts since such contacts may provide knowledge on foreign conditions. There are different channels through which foreign contacts can take place, the most important is through the foreign direct i nvestment (FDI) by multinational enterprises (MNEs).

MNEs typically have a presence in many markets, making them a potential source of information about foreign markets, consumers and technology. Therefore, higher foreign equity participation by MNEs may lead to higher export performance. This impact on exports of domestic firms is through direct contact with the multinationals (a direct effect). Sometimes, the presence of MNEs in the domestic market itself would increase the export performance of domestic firms (an indirect effect). The information with the MNEs on foreign markets may leak out to the domestic firms even if they do not participate in joint ventures with foreign firms. This externality is one type of “spillovers” from FDI. S pillovers can also take place when the presence of MNEs i mproves the productive efficiencies of domestic firms, making their products competitive in price and quality in the international market, thus improving their export performance. These types of spillovers are known as “horizontal spillovers” since they occur to domestic firms in the same industry group of foreign firms through competition. Similarly, the multinational buyers of intermediate goods may provide information about other possible international purchasers of those products, so that the domestic intermediate goods producers can expand their production and achieve the economies of scale that will in fact reduce the price of their p roducts. This aspect of spillovers from foreign

Economic & Political Weekly

EPW
December 26, 2009 vol xliv no 52

firms arising through buyer-supplier linkages are mostly known as “backward spillovers”.

The literature classifies FDI based on MNE’s market strategy into domestic market-seeking and export-oriented. Most of the developing countries (China, Malaysia, Indonesia, Thailand, etc) are now mostly looking at export-oriented FDI to strengthen their export competitiveness. Pradhan and Abraham (2005) argue that export-oriented FDI can be expected to generate strong links with the local economy compared to local market-oriented FDI in the host country specifically because it is motivated to exploit the l ocation advantages offered by the host country such as low-cost labour, raw materials, components and parts, among others. The presence of export-oriented FDI and the interaction with them may induce the domestic firms to diversify into export market when information on foreign markets brought in by foreign firms spill over to them. While local market-oriented FDI may crowd out domestic firms and investments, export-oriented FDI can stimulate investment by generating demands for intermediate goods (Pradhan and Abraham 2005). The experience of China and Mexico shows that these countries were able to increase their international market shares mostly by attracting export-oriented FDI (UNCTAD 2002). Therefore, it is expected that spillover effects will be larger from the presence of export-oriented MNEs compared to domestic market-seeking ones.

However, such type of growth-led, efficiency seeking and e xport-oriented FDI can be materialised only if the regulatory regime facilitates greater freedom to the MNEs in their operations (Athreye and Kapur 2001; Aggarwal 2002). According to Aggarwal (2002), the protected markets are more likely to attract tariffjumping FDI which are, therefore, likely to be more domestic marketseeking. Economic liberalisation through opening up of the economy will force the tariff-jumping, market-seeking type of MNEs to restructure their strategies due to increased competition from imports, new foreign firms and diligent indigenous firms. The existing MNE-affiliates would have to look at technology upgradation either through technology imports or through research and development (R&D) activities to strengthen their competitiveness. The MNEs would also have to look at external markets to maintain their growth and profitability. Therefore, economic l ibera lisation would not only attract more efficiency-seeking, e xport-oriented FDI but also force the existing MNEs to reframe their strategies.

Based on the above argument, this study hypothesises that the presence of foreign firms with their sales, international marketing exposure and expertise are expected to increase the competitiveness and export orientation of domestically owned firms in developing countries through either “horizontal spillovers” or “backward spillovers”. This hypothesis is examined using firm level data for Indian manufacturing industry for the period 1993 to 2008 (a period after the economic liberalisation of 1991).

India’s Trade and FDI Policies

India’s economic policy after independence was one of national self-sufficiency that stressed the importance of government regulation of the economy and planned industrialisation. Its economic policy was considered as inward looking and highly interventionist. The salient features of the policy included import protection, complex industrial licensing requirements and substantial public ownership of industries, especially heavy industries, among

o thers (Topalova 2004). India’s trade policy was characterised by high tariffs and pervasive import restrictions. These restrictive policies continued till the 1970s. However, amidst growing dissatisfaction over its results, there was a gradual shift in the focus of India’s development strategy towards export-led growth during the 1980s. Seeing the experience of many east Asian countries in achieving high growth and poverty reduction through policies that emphasised greater export orientation and encouragement of the private sector, several liberalisation measures were adopted during the 1980s and 1990s. Industrial licensing was eased and import restrictions were brought down. The 1991 liberalisation included major structural reforms including trade and foreign investment liberalisation. Realising the importance of FDI through MNEs in improving productivity, exports and overall economic growth, a number of policy decisions were undertaken to attract more FDI. This has resulted in increased inflow of FDI. The FDI inflow has increased from a mere $97 million in 1990-91 to $34,362 million in 2007-08 (RBI 2009).

Along with the foreign investment policies, there was tremendous reduction in tariffs and non-tariff barriers and also in the protection to Indian industries (Ahluwalia 2002).1 The consistent trade reform policies ushered in through various export-import policy plans helped India to increase its share in world exports from 0.5% in the first half of the 1990s to 0.7% in 2000-01, and to 1% in 2005-06. This share in world exports remained the same at 1% till 2007-08. The ratio of total exports to the gross domestic product (GDP) rose from 5.8% in 1990-91 to 12.2% in 2004-05, and grew further to 14% in 2006-07 (GoI 2008). Merchandise exports from India increased rapidly from $18,145 million in 1990-91 to $162,983.90 million in 2007-08, with almost 70% of the export contribution coming from the manufacturing sector (GoI 2009).

Review of Literature

The role of FDI in the export performance of host country industries has received considerable attention in recent years, especially in the context of liberalisation and globalisation. A cross-country analysis of 52 countries by the UNCTAD (1999) found that there is a positive relationship between FDI and manufactured exports, and the relationship is stronger for developing countries than for d eveloped countries and in high and low-tech industries than in medium-tech ones. Aitken et al (1997) conducted a study on Mexican manufacturing firms for the period 1986-90 and found that export decision of Mexican firms is positively related to the presence of foreign firms; which is measured using two separate variables – MNEs’ production and their exports. They found that the presence of MNEs with their production and export activities positively influence the export performance of Mexican firms. Kokko et al (2001) examined the association between FDI spillovers and the export behaviour of domestic firms in Uruguay using a crosssectional firm level data. They found that domestic firms are more likely to export if they operate in sectors where the presence of foreign firms is relatively high. Their study also pointed out that the type of trade regime (controlled or liberalised) may influence the ability of MNEs in generating positive e xport spillovers.

December 26, 2009 vol xliv no 52

EPW
Economic & Political Weekly

Sjoholm (1999) examined the various types of foreign contacts that influenced as well as enabled the establishments to become exporters in the Indonesian manufacturing sector. Taking three types of foreign contacts – foreign ownership, imports, and spillovers from regional presence of FDI – the study found that while the first two types of foreign contacts (ownership and imports) have a positive effect on the propensity to become an exporter, there are no export spillovers from a large regional presence of FDI. Greenaway et al (2004), using a two-step Heckman selection model2 to determine the influence of FDI spillovers on the export decision of domestic firms, found positive FDI spillovers on the probability of a United Kingdom firm being an exporter. They found that the most important channel of export spillovers is the increased competition resulting from foreign firms.

A number of studies have attempted to analyse the impact of FDI on the export performance of Indian industries. In India, earlier studies for the period of restrictive policy regimes could not find any significant difference in the export performance of foreign and domestic firms (Kumar and Siddharthan 1994). However, a number of studies for the post-1991 liberalisation period suggest that foreign firms have shown significantly higher export performance as compared to domestic firms (Aggarwal 2002; Kumar and Pradhan 2003). Aggarwal (2002) compared the export performance of MNE affiliates and domestic firms in Indian manufacturing after the 1991 liberalisation by analysing the determinants of their export intensities. The study examined the relationship between FDI and export performance using the Tobit model for 916 Indian manufacturing firms for the period 1996-2000. Aggarwal found that the liberalisation measures of the 1990s enhanced the export role of MNE affiliates, especially in the late 1990s. However, she could not find any evidence of a positive relationship between f oreign equity share and export performance of firms.

Kumar and Pradhan (2003) looked at the important factors that influence the export competitiveness of Indian manufacturing firms with an emphasis on knowledge-based industries. They found that younger firms drive export competitiveness in the high and low technology industries whereas older firms are more competitive in the medium technology industries. The study also found that foreign affiliates are better achievers on the export front compared to their domestic counterparts in Indian manufacturing. The study concluded that the liberalisation policies of the 1990s have definitely improved the export competitiveness of Indian manufacturing, especially in the technology-intensive segments. Banga (2003) also found a significant impact of FDI on the export intensity of non-traditional export industries in India.

Most of these studies examined the role of foreign equity participation in the export decision of firms in Indian manufacturing and merely compared the export performance of domestic and foreign firms without looking into the possibilities of export s pillovers from foreign firms. This study attempts to analyse the effect of FDI spillovers on the export performance of Indian firms in a liberalised framework.

Methodology and Data

This study uses appropriate econometric models to estimate the effect of FDI spillovers on firm-level export performance. It should

Economic & Political Weekly

EPW
December 26, 2009 vol xliv no 52

be noted that it is not easy to separate the influence of FDI spillovers on the export performance of a firm since the export behaviour of a firm is not a simple function of the foreign presence. Many other firm-specific or industry-specific factors do exert a deterministic influence on the firm-level export performance. Following the earlier literature, this study also therefore, tries to incorporate some of the other important factors (or controlled variables) such as firm’s size, capital intensity, R&D intensity and technology import intensity on the export behaviour of the firm. The explanation and construction of variables used in the exportdetermination model is discussed below.

Variables

Age: The age of the firm is considered as an important determinant of its export behaviour since it may act as a proxy for the a ccumulated knowledge and skill in production and marketing. Firms with longer experience in the market are supposed to perform relatively better on the export front (Sjoholm 1999; Kumar and Pradhan 2003). However, this study does not use the age of the firm in years but instead incorporates a dummy variable to demarcate the export behaviour of firms incorporated before the major export promotion policies of 1983 and firms incorporated thereafter.

Firm Size: The size of the firm is expected to positively influence the export performance of the firm because of its significant economies of scale (Caves 1996; Ruane and Sutherland 2004). Large firms are relatively likely to export more because they enjoy economies of scale due to their size which may make their products more price-competitive in the export market. Ruane and Sutherland point out that relatively larger enterprises are more capable of absorbing any fixed costs associated with entering into an export market and exploiting economies of scale in the exporting process. Another argument is that large establishments have been relatively successful in the domestic market, which could increase the possibility to succeed internationally (Sjoholm 1999).

Capital Intensity: Capital intensity is supposed to have positive effects on the quality of the products, and perhaps on the ability to export (Sjoholm 1999). However, Kumar and Siddharthan (1994) consider that a higher capital intensity of operations is unlikely to give the firm a competitive advantage in a developing country setting as in India with an abundance of labour and a relative scarcity of capital.

Technology: Technology intensity is supposed to capture the quality of the product, which is a major factor for export competency. Technology can be either from own R&D activities or through imports. Again, technology imports have two facets – embodied and disembodied. If the technology acquisition is through capital goods imports, then technology is assumed to be embodied in the machineries and equipments imported. On the other hand, technology licensing by paying royalties and lump sum payments are considered as disembodied technology imports. It is assumed that firms with a higher R&D spending may absorb information externalities related to exporting efficiently. It is also postulated that firms with high technology imports both through embodied (import of capital goods) and disembodied (technology licensing or purchasing) channels, would have a better export performance. Access to imported foreign technology helps the firms to produce goods that may well suit foreign markets, making them more competitive. Therefore, such technology imports are supposed to positively influence the export performance of firms.

FDI Spillovers: In order to understand the role of knowledge s pillovers from foreign firms on the export performance of domestic firms, two possible spillover measures are included in the basic estimation equation. The first spillover measure used is the spillover from sales of foreign firms. The market share of foreign firms in the form of their sales can affect the productivity and exports of domestic firms to a great extent. The other spillover measure is the spillover from exports of foreign firms. Foreign firms are generally considered as more export oriented because of their marketing expertise and experience along with worldwide marketing networks. The presence of such foreign firms may influence the domestic firms to be more competent in accessing the international markets. These spillover measures are used to construct the “horizontal” (capturing the externalities due to competition) and “backward” (due to buyer-supplier linkages) spillover variables used in the study.

The two spillover measures (spillover from exports and spillover from sales) used in the study are not mutually exclusive always and we find significant correlation between these variables in this study. Therefore, we examine the impact of these spillovers, incorporating them separately in the estimation equation.

Model Specification

For a given period t, EXPINTij = β0 + β1 Sizeij + β2 Ageij + β3 CAPINTij

+ β4 RADINTij + β5TIMINTij +β6Horiij + β7Backij + εij (1) where the subscripts i and j denote firm and industry, respectively, and, EXPINT (Export Intensity) = Exports/Sales SIZE (Firm Size) = ln (Sales) AGE (A dummy variable) = 1, if the firm was incorporated before 1983 (the starting year of export promotion policies in India)

= 0, otherwise CAPINT (Capital Intensity) = Net fixed assets/sales RADINT (R&D Intensity) = R&D expenditure/sales TIMINT (Technology Import Intensity) = (Capital Goods Import (Embodied Technology) + Technology Licensing (Disembodied Technology))/sales. Hori = Horizontal Spillovers from foreign firms’ exports (EXHori) or sales (SAHori). Back = Backward Spillovers from foreign firms’ export (EXBack) or sales (SABack) in other industry sectors.

Construction of FDI Spillover Variables

Horizontal Spillovers (Hori): The horizontal spillovers from the presence of foreign firms in the same industry sector is constructed by taking the share of foreign firms’ sales in a particular sector (say, j) to the total sales of that sector for a given time p eriod t:

SALES FOREIGN, j

Hori j = (2a)

SALES TOTAL, j

Similarly another measure of spillovers is defined by using exports ratio, i e, by taking the share of foreign firms’ exports in a particular industry sector (say, j) to the total exports of that s ector at time period t

EXPORTS FOREIGN, j

Hori j = (2b)

EXPORTS TOTAL, j

It is expected that firm-specific characteristics such as firm size may influence the extent of spillovers from foreign firms to the domestic firms. The spillover variables are therefore multiplied by the market share of each domestic firm in their concerned i ndustry-sector to control for such firm-specific determinant

f actors.
Hori ij = SALESij SALES j * SALES FOREIGN, j SALES TOTAL , j (3a)
Hori ij = SALESij SALES j EXPORTS FOREIGN , j * EXPORTS TOTAL , j (3b)
Spillovers from Outsourcing or Backward Linkages (Back):

Following Smarzynska (2004), the spillovers through outsourcing (or backward linkages) for a given time period t is computed as

n

Back j = ∑αjh Hori h (4)

h=1 h≠ j

where αjh is the proportion of the output of sector j supplied to sector h, taken from the Input-Output (IO) absorption matrix (1998-99 and 2003-04)3 and n is the number of industry sectors considered for the study. The greater the proportion of output to an industry sector j, with larger foreign presence from sector h, then the greater will be the backward spillovers from sector h to sector j. Here we consider only the spillovers due to backward/ supplier linkages between firms in different industry sectors.

To control for the firm-specific characteristics that determine the extent of spillovers, equation (4) is multiplied by the market share of each firm.

SALES ij n

Backij = ∑ Hori h (5)

* αjh

h=1SALESj h≠ j

In order to control for industry specific factors that may influence the export intensity of firms, industry dummies are introduced in the model. The sample firms for the study are grouped into eight two-digit industry sectors (the details of industry sectors are given in the section on data). Therefore, seven industry dummies, represented by Di, are introduced in equation (1), taking food processing as the reference industry.

7

expintij = β0+∑β0i Di +β1 sizeij + β2 ageij + β3 capintij +

i=1

(6) β4 radintij + β5 timintij + β6 Horiij + β7 Backij + εij

The data for the study is taken from the Prowess database (an electronic database for Indian corporate firms provided by the

December 26, 2009 vol xliv no 52 EPW Economic & Political Weekly

Centre for Monitoring Indian Economy (CMIE)). Manufacturing e stablishments from 35 three-digit National Industry Classification (NIC) industries are taken for this study for the period 1993-2008. The sample firms cover a wide range of manufacturing i ndustries that include food processing, textiles, chemical, non-metallic minerals, metal and metal products, basic machinery, electrical, electronics and medical equipments, and automobiles.

To understand the export spillovers from FDI, sample firms in each industry sector have been classified into two groups: domestic controlled/owned firms (referred as domestic firms) and foreign controlled/owned firms (referred as foreign firms). The criteria used for this classification is the one given by the Reserve Bank of India (RBI), which considers firms with foreign equity participation of 10% or more as foreign controlled/owned firms.

In order to identify the role of foreign presence (spillovers) – either with their sales or exports – on the export performance of the domestically-owned firms, the whole analysis is separated into four time periods to understand the export performance of firms over different phases of liberalisation, viz, 1993-96, 1997-2000, 2001-04, 2005-08. The average values of variables are used for each of these four time periods to take care of the missing data and also the time specific factors that may influence the export intensities of firms. The industry-wise classification of sample firms for the four time periods is given in Table 1. From Table 1 it should be noted that the share of domestic and foreign firms in the sample has remained almost the same in all the i ndustry groups and for the four time periods.

Performance Comparison of Domestic and Foreign Firms

The performance comparisons of both domestic and foreign firms in terms of indicators such as export intensity, technology intensities through own R&D as well as through technology imports, and both labour and capital productivities for different time periods under

Table 1: Industry-wise Classification of Sample Firms

the study are provided in Table 2. To examine whether there are any significant changes in the average intensities of both domestic and foreign firms over the period of time, the usual t-statistic (or normalised z-statistic) is used after examining the equality of variances.4 These results are provided in Appendix 1 (p 105).

From Table 2, we note that the average export intensity of domestic firms has increased significantly from 10.43% during 1993-96 to 12.34% during 1997-2000. However, though this has increased to 13.48% during 2001-04, and further to 15.02% during 2005-08, these changes are not significant. The increase in the average export intensity of foreign firms was around 11.65% during 1993-96 which rose to 13.18% during 1997-2000 but fell to 12.95% during 2001-04. However, even these changes are not significant. Compared to earlier periods, there was a significant i ncrease in the average export intensity of firms during the period 2005-08, which has increased significantly from 12.95% in 2001-04 to the level of 16.34% during 2005-08. Our analysis r eveals that there are no significant differences in the export i ntensities of domestic and foreign firms in all the periods.

The R&D spending by the foreign firms as a share of their sales is found to be significantly higher compared to that of domestic firms only during 1997-2000. Table 2 shows that there was a steady increase in the R&D intensities of domestic firms from an average of 0.23% during 1993-96 to 0.25% during 1997-2000, then to 0.31% during 2001-04, and further to 0.78% during 2005-08. However, none of these changes are significant in any of these time periods except between 2001-04 and 2005-08 where the average R&D intensities of domestic firms has increased significantly (at 8% significance level). In the case of the average R&D intensity of foreign firms also the periodic changes are not significant. It should be noted that the average R&D intensity of foreign firms has come down from 0.42% during 1997-2000 to 0.35% during the period 2001-04. The analysis shows that there is no clear trend in the aver

age R&D intensity of foreign

Industry Sector 1993-96 1997-2000 2001-04 2005-08 firms during the four time All (No) Domestic (%) Foreign (%) All (No) Domestic (%) Foreign (%) All (No) Domestic (%) Foreign (%) All (No) Domestic (%) Foreign (%)

periods under study. How-

Food products 159 (86.16) (13.84) 155 (86.45) (13.55) 167 (85.63) (14.37) 226 (89.38) (10.62)

ever, the increasing trend in

Textiles 202 (91.58) (8.42) 229 (91.70) (8.30) 236 (91.95) (8.05) 257 (92.22) (7.78)

the average R&D intensity of

domestic firms between

Non-metallic minerals 87 (82.76) (17.24) 89 (80.90) (19.10) 86 (81.40) (18.60) 113 (78.76) (21.24)

2001-04 and 2005-08 sug-

Metal and metal products 200 (89.50) (10.50) 202 (87.13) (12.87) 195 (88.72) (11.28) 257 (85.60) (14.40)

Chemicals 438 (84.25) (15.75) 487 (84.80) (15.20) 473 (85.20) (14.80) 484 (81.40) (18.60)

Basic machinery 132 (63.64) (36.36) 135 (63.70) (36.30) 154 (67.53) (32.47) 174 (68.97) (31.03) gests that the liberalisation
Electrical and electronics 146 (76.71) (23.29) 165 (76.97) (23.03) 161 (77.02) (22.98) 169 (76.33) (23.67) policies and the consequent
Automobiles 102 (64.71) (35.29) 104 (66.35) (33.65) 102 (65.69) (34.31) 150 (70.00) (30.00) entry of foreign firms are
Total 1,466 (82.13) (17.87) 1,566 Figures in parentheses are row-wise percentage for each period. (82.18) (17.82) 1,574 (82.66) (17.34) 1,830 (81.75) (18.25) compelling the dome stic firms to invest more in R&D
Table 2: Performance Comparison of Domestic and Foreign Firms Performance Indicators 1993-96 1997-2000 Domestic Foreign Difference Domestic Foreign Difference Domestic 2001-04 Foreign Difference Domestic 2005-08 Foreign Difference to improve their competitiveness both in the domestic
Average export intensity Average R&D intensity 0.1043 0.0023 0.1165 0.0030 -0.0122 0.1234 0.1318 -0.0007 0.0025 0.0042 -0.0084 -0.0017* 0.1348 0.0031 0.1295 0.0035 0.0053 -0.0004 0.1502 0.0078 0.1634 0.0047 -0.0132 0.0031 and international markets. The average technology
Average technology imports as a share of total
import intensity Average value added per labour 0.0508 0.0533 -0.0025 0.0124 0.0252 -0.0129** 6.1800 4.4304 1.7496** 4.5617 3.2779 1.2838** 0.0070 0.0121 -0.0051** 2.9566 2.7750 0.1816 0.0157 N A 0.0227 -0.0070** N A N A sales is signi ficantly higher for foreign firms as compared
Average value added to domestic firms in all the
per capital 0.5336 0.6452 -0.1117* 0.3603 0.5356 -0.1753** 0.2408 0.4619 -0.2212* N A N A N A time periods except during
N 1,204 Significant at ** 1% and * 5% level. Source: Computed from sample data. 262 1,287 279 1,301 273 1,496 334 the first phase, i e, during 1993-96 (Table 2). This is
Economic & Political Weekly December 26, 2009 vol xliv no 52 101
EPW

e xpected because foreign firms mostly depend on their parent com-foreign firms during the period 2001-04. Contrary to this, the expany (or foreign collaborator in case of joint ventures) for technology port intensities of domestic firms show an upward trend in almost which they may bring in either through the imports of capital goods all these industry sectors. This shows that the economic where the technology is embodied or through licensing agreements l iberalisation and globalisation policies in India after 1990, apart with the foreign collaborator. However, it should be noted that the from other things, might have helped to improve the export oriendependence on imported t echnology, as measured from the ratio of tation of the domestic firms partly because of the external e xposure technology imports to total sales, is declining for both domestic and and partly because of the increased competitiveness. However, the foreign firms after the 1991 liberalisation. declining trend in the export intensities of foreign firms during the

Labour productivity (measured

Table 3: Spillovers from Foreign Firms’ Exports on the Export Performance of Domestic Firms (Results from Robust Regression Analysis)

as value added per unit of labour) Dependent Variable EXPINT Without Industry Dummies With Industry Dummies
is higher for the domestic firms when compared to foreign firms in the first three time periods and SIZEAGE 1993-96 0.0033** (0.0006) 0.0066** 1997-2000 0.0043** (0.0008) 0.0061** 2001-04 0.0056** (0.0009) 0.0035 2005-08 0.0063** (0.0013) 0.0032 1993-96 0.0037** (0.0006) 0.0068** 1997-2000 0.0046** (0.0008) 0.0068** 2001-04 0.0063** (0.0010) 0.0043 2005-08 0.0096**(0.0015) 0.0027
these differences are significant (0.0018) (0.0024) (0.0028) (0.0034) (0.0019) (0.0025) (0.0030) (0.0038)
at least in the first two periods CAPINT -0.0011 0.0015 -0.0003 -0.0012 -0.0007 0.0016 -0.0003 -0.0012
(1993-96 and 1997-2000) (Table 2). However, labour productivity shows a significant declining trend RADINTTIMINT (0.0013) 0.2567** (0.0674) 0.0305** (0.0018) 1.6531** (0.1575) 0.0456 (0.0008) 0.3578** (0.1132) 0.1768** (0.0011) 0.3068** (0.0239) 0.8751** (0.0013) 0.2590** (0.0712) 0.0291** (0.0019) 1.7107** (0.1679) 0.0316 (0.0008) 0.4000** (0.1220) 0.1806** (0.0013) 0.3113**(0.0265) 0.8069**
in both domestic and foreign (0.0045) (0.0302) (0.0544) (0.0418) (0.0048) (0.0320) (0.0583) (0.0483)
firms for all the periods during EXHori 0.0033 -0.0039 0.0091 0.0028 0.0067 0.0018 0.0191 0.0070
1993-2004. A similar type of de- EXBack 0.0044) 0.2121 (0.0060) 0.0966 (0.0079) 0.1006 (0.0115) -0.2387 (0.0057) 0.2314 (0.0077) 0.1418 (0.0103) 0.0231 (0.0169) -0.7301
clining trend is also observed in the (0.1275) (0.1463) (0.1868) (0.3254) (0.1388) (0.1620) (0.2130) (0.4048)
case of capital productivity (meas- Textiles 0.0154** 0.0197** 0.0262** 0.0453**
ured as value added per unit of (0.0036) (0.0046) (0.0057) (0.0072)
capital). Unlike in the case of labour productivity, capital produc- Chemical Non-metallic mineral 0.0077** (0.0031) 0.0073 0.0106** (0.0040) 0.0025 0.0183** (0.0051) 0.0034 0.0241** (0.0065) -0.0114
tivity is higher for foreign firms (0.0045) (0.0062) (0.0076) (0.0098)
when compared to domestic firms Metal products 0.0073* 0.0127** 0.0249** 0.0086
in all the first three time periods for (0.0037) (0.0048) (0.0059) (0.0078)
which data is available. This higher performance of foreign firms in Basic machinery Electrical and electronics 0.0087* (0.0040) 0.0037 0.0102 (0.0053) 0.0028 0.0269** (0.0068) 0.0113 0.0228** (0.0093) 0.0195
comparison to the performance of (0.0039) (0.0054) (0.0068) (0.0104)

domestic firms is significant in all Automobile 0.0152** 0.0140* 0.0095 0.0122 (0.0045) (0.0063) (0.0082) (0.0098)

these three time periods. This in-

Constant -0.0002 -0.0001 0.0001 -0.0022 -0.0093* -0.0108* -0.0189** -0.0289**

formation supports the a ssumption

(0.0025) (0.0037) (0.0041) (0.0072) (0.0039) (0.0051) (0.0063) (0.0093) that in a country like India, where Observations 1203 1286 1301 1495 1203 1286 1301 1495

labour supply is large and the gov-F-ratio 21.55 27.22 12.00 93.46 12.46 15.42 8.65 43.53

ernment policies support labour-R2 0.112 0.130 0.061 0.306 0.128 0.145 0.086 0.292# Significant at ** 1% and * 5% level.

intensive production, the domestic # The lower R2 value in this case compared to the R2 value without industry dummies may be due to the differences in the pseudo values of the dependent variable created and used as response variable while using the “rreg” programme in STATA.

firms tend to be more labour-inten-

Figures in parentheses are standard errors.

sive, while foreign firms, mostly from the developed world, may be more capital-intensive.

Comparison of Export Intensities

The export intensities of domestic and foreign firms in different industry sectors for the period 1993 to 2008 are provided in F igure 1 (p 104). The differences in the export intensities of these two groups were examined statistically using annual data (refer A ppendix 2 (p 105) for the results). It is found that the export intensities of domestic firms are significantly higher in chemicals industry sector when compared to foreign firms and the same in auto mobiles and food products sectors, whereas foreign firms have significantly higher export intensities in all the other sectors.

Another major point to be noted (Figure 1) is that in food products, textiles, chemicals, basic machinery, and electrical and e lectronics sectors, there was a fall in the export intensities of the period 2001-04 in sectors like food products, chemicals, and electrical and electronics indicates that these firms were focusing their attention towards the Indian domestic m arket rather than the international market for their sales. However, the figure shows that during the period 2005-08, in almost all the industry sectors except food processing and automobile sectors, foreign firms demonstrated an increased trend in their export orientation. A major reason may be identification of India as a major export-hub by foreign collaborators in these i ndustry s ectors.

Estimation Procedure and Results

Equations (1) and (6) are estimated using Ordinary Least Squares (OLS) estimation method but found heteroskedasticity in all the time periods. A regression diagnostic test also showed the presence of influential outliers5 with high leverages6 in the data and these

December 26, 2009 vol xliv no 52

EPW
Economic & Political Weekly

outliers would result in imprecise prediction and distort the signifi-Table 2 indicates that the economic liberalisation in India has not cance of parameter estimates. Therefore, a robust regression esti-attracted much export-oriented FDI. mation technique that would dampen the i nfluence of outliers is The estimation results, using the foreign firms’ sales to construct used to take care of the above mentioned drawbacks (Montgomery the spillover variables, are reported in Table 4. Contrary to the et al 2004). The “rreg” programme in STATA package is used to esti-findings regarding horizontal spillovers from foreign firms’ exmate the regression equations, where the m edian regression ports, the analysis shows that there are significant spillover e ffects method has been used to take care of the o utliers. from the sales of foreign firms on the export performance of

Tables 3 (p 102) and 4 provide the estimation results for the d omestic firms in the same industry group. This is found signifioverall manufacturing industry. Estimation results are reported cant in almost all time periods, except during 1997-2000 and for the four time periods (1993-96, 1997-2000, 2001-04 and 2005-08 when the industry characteristics are not controlled for. 2005-08) with and without industry dummy variables. Hetero-However, once the industry characteristics are controlled using skedasticity is found in the usual OLS estimation for all the i ndustry dummies, the coefficient of the horizontal spillovers periods. As discussed earlier, in order to take care of hetero-(SAHori) from foreign firms’ sales is significant and positive in all skedasticity as well as the influence of outliers, robust regression the periods, except during 1997-2000. This indicates that the commethod with median regression estimation is used. Table 3 petition from the presence of foreign firms with their sales in the provides the estimation results for the overall manufacturing, domestic market is forcing the domestic firms to look for markets where foreign firms’ exports share is taken to construct the spill-abroad. As in the case of spillovers from foreign firms’ e xports over variables, EXHori and EXBack. These two main variables are (EXBack), the variable representing backward spillovers from used to capture the horizontal and backward spillovers from for-f oreign firms’ sales (SABack) does not indicate significant impact eign firms’ exports, respectively. The horizontal spillover variable on domestic firm’s export performance in all the time p eriods. (EXHori) is not significant in any of the model specifications for The coefficient of the SIZE variable is positive and significant in all the four time periods. Similarly, the backward spillover varia-all the model specifications and time periods, supporting that larger ble (EXBack) is also not significant in all the four time periods firms in terms of sales are more export-intensive. This indicates that under study. This along with the analyses of average export the economies of scale achieved through larger size and the indusintensities of domestic and foreign firms already explained in try dominance of firms have a major influence on their export

performance. The previous

Table 4: Spillovers from Foreign Firms’ Sales on the Export Performance of Domestic Firms (Results from Robust Regression Analysis)

studies (Kumar and Sid-

Dependent Variable EXPINT Without Industry Dummies

With Industry Dummies 1993-96 1997-2000 2001-04 2005-08

1993-96 1997-2000 2001-04 2005-08 dharthan 1994; Aggarwal

SIZE 0.0036** 0.0043** 0.0057** 0.0064** 0.0041** 0.0048** 0.0066** 0.0097**2002; Kumar and Pradhan (0.0006) (0.0008) (0.0009) (0.0013) (0.0007) (0.0009) (0.0010) (0.0015)

2003) found firm size as an

AGE 0.0066** 0.0059** 0.0037 0.0030 0.0067** 0.0067** 0.0043 0.0022

important determinant of ex

(0.0019) (0.0023) (0.0028) (0.0035) (0.0020) (0.0025) (0.0030) (0.0038) CAPINT -0.0009 0.0015 -0.0003 -0.0012 -0.0004 0.0015 -0.0002 0.0012port intensity of firms. The (0.0013) (0.0018) (0.0008) (0.0012) (0.0014) (0.0019) (0.0008) -(0.0013)

coefficient of AGE variable is

RADINT 0.2556** 1.6038** 0.3363** 0.3073** 0.2616** 1.6611** 0.3797** 0.3130**

also significant with positive

(0.0686) (0.1574) (0.1148) (0.0241) (0.0737) (0.1707) (0.1232) (0.0262) TIMINT 0.0299** 0.0461 0.1805** 0.8785** 0.0282** 0.0353 0.1818** 0.8071**sign in the first two time pe (0.0046) (0.0298) (0.0549) (0.0421) (0.0050) (0.0323) (0.0589) (0.0476)

riods both before and after

SAHori 0.0115** 0.0032 0.0167* 0.0140 0.0170** 0.0128 0.0364** 0.0409*

contro lling for industry ef

(0.0045) (0.0063) (0.0085) (0.0123) (0.0060) (0.0083) (0.0112) (0.0168) SABack 0.2730 0.1180 0.0785 -0.3333 0.3080 0.1696 0.0258 -0.9874fects. This dummy variable (0.1834) (0.1702) (0.2070) (0.4710) (0.2033) (0.1939) -(0.2363) (0.5685)

indicates that the manufac-

Textiles 0.0171** 0.0217** 0.0302** 0.0490**

turing firms incorporated be

(0.0036) (0.0046) (0.0059) (0.0073) Chemical 0.0069* 0.0109** 0.0185** 0.0215** fore the e xport-oriented poli(0.0031) (0.0040) (0.0051) (0.0064)

cies of 1983 are superior in

Non-metallic mineral 0.0078 0.0034 0.0067 -0.0135

their export performance

(0.0046) (0.0062) (0.0077) (0.0094) Metal products 0.0082* 0.0139** 0.0268** 0.0088 compared to firms incorpo(0.0036) (0.0048) (0.0059) (0.0071)

rated after that period, at

Basic machinery 0.0094* 0.0108* 0.0293** 0.0230**

least in the initial years of lib

(0.0041) (0.0053) (0.0068) (0.0086) Electrical and electronics 0.0021 0.0019 0.0102 0.0166 eralisation. It indicates that (0.0041) (0.0054) (0.0068) (0.0089)

the accumulated learning of

Automobile 0.0116* 0.0109 0.0049 0.0043

firms had a significant influ

(0.0049) (0.0066) (0.0083) (0.0095) Constant -0.0024 -0.0020 -0.0017 -0.0045 -0.0118** -0.0137** -0.0243** -0.0356**ence on their export perform (0.0026) (0.0037) (0.0043) (0.0073) (0.0040) (0.0053) (0.0066) (0.0096)

ance at least for a certain pe-

Observations 1203 1286 1301 1495 1203 1286 1301 1495

riod even a fter the full-scale

F-ratio 22.26 27.26 12.30 91.99 12.57 15.32 9.22 44.69

liberalisation policies of 1991

R2 0.115 0.130 0.062 0.302 0.129 0.144 0.091 0.297#

that were supposed to en-

Significant at ** 1% and * 5% level. # The lower R2 value in this case compared to the R2 value without industry dummies may be due to the differences in the pseudo values of the

courage the setting up of

dependent variable created and used as response variable while using the “rreg” programme in STATA. Figures in parentheses are standard errors. new export-oriented firms.

Economic & Political Weekly

EPW
December 26, 2009 vol xliv no 52

Foreign Domestic All

However, during the last two time periods (2001-04 and 2005-08) the AGE variable do not show significant influence on the export performance of firms. The capital intensity (CAPINT) variable is not significant in any of the time periods and in any of the model specifications. This implies that in India still the labour-intensive products are capturing the export market. The technology factors that may influence the export performance of firms are examined using R&D intensity (RADINT) and technology import intensity (TIMINT) of firms. The coefficients of both these variables are positive and

Figure 1: Export Intensity of Domestic and Foreign Firms – Industry-wise (%)

Food Products

14

Foreign All Domestic
60 12 50

10 40

8 30

6 20

4 10

1993 1996 1999 2002 2005 2008 1993 1996

Chemicals Non-Metallic Minerals

20

All Foreign Domestic
9 18 8

16

7 14

6 12

5 10

4

8 3

6 2

4

1993 1996 1999 2002 2005 2008 1993 1996

Metal and Metal Products 40

Domestic Foreign All
18 1635 1430 12

25 10

20 8

15 6

10

4 5

2 0

0

1993 1996 1999 2002 2005 2008 1993 1996

Electrical and Electronics 12

16 14

Foreign Domestic All
10 12

8

10 8 6

6 4

4

2 2

0 01993 1996 1999 2002 2005 2008 1993 1996

Source: Computed from sample data.

s ignificant in almost all the time periods and model specifications.7 The significant positive coefficients for RADINT suggest that firms’ own technology effort is an important factor in c ontributing to their competitiveness in the international market. The coefficient of technology import intensity variable also has a significant positive sign for all the periods. Technology licensing (disembodied technology imports) and imports of capital goods like machineries where

104 t echnology is embodied, are two ways of acquiring foreign knowledge for developing countries like India where indigenous capabilities are limited due to financial and scientific resource constraints (Kumar and Pradhan 2003; Joseph 2005). The results suggest that such type of technology imports have helped in improving the e xport competitiveness of Indian firms.

Foreign Domestic All
Domestic Foreign All

The industry-specific dummy variables show that in comparison to firms in food products industry, firms in textiles, chemicals, and basic machinery are more export-oriented in almost all the four

Textiles

Foreign All Domestic

1999 2002 2005 2008

1999 2002 2005 2008 Basic Machinery

1999 2002 2005 2008 Automobiles

1999 2002 2005 2008

performance of domestic firms in the same industry group. This indicates that the increased competition in the domestic market after the economic liberalisation, through sales of foreign firms, is forcing the domestic firms to be more competitive and look for markets abroad.

The study finds that the firm size and technology factors, either through own R&D activities or through technology imports,

December 26, 2009 vol xliv no 52

EPW
Economic & Political Weekly

time periods. Metal and metal products firms in the first three periods and automobile firms in the first two periods show significantly higher export performance compared to food products firms. The export intensities of domestic firms in electrical and electronics industry are not significantly different from that of domestic firms in food products industry in all the four time periods under study.

Conclusions

Our results do not provide any clear support to the hypothesis that there exist spillovers (either horizontal or backward spillovers) from the presence of foreign firms to influence the export performance of domestic firms. Sjoholm (1999) also found that increased foreign presence does not seem to benefit export (i e, export spillovers from f oreign firms are not significant) in Indonesian manufacturing firms. As discussed earlier, the spillover effects may not be much relevant to domestic exporters if the MNEs in the sector concentrate more on the domestic market. Though the study does not find significant horizontal spillovers from foreign firms’ exports, the analysis indicates that there are significant spillover effects from the sales of foreign firms on the export d etermine the export intensities of domestic firms in the Indian manufacturing sector. This implies that larger firms are able to exploit export markets because of economies of scale. Technological capabilities obtained through own R&D and technology imports are important for achieving export competitiveness, e specially in the globalisation era.

Economic liberalisation is expected to attract more efficiencyseeking and export-oriented FDI. However, our findings support the argument that the MNEs in India have not yet started to take the comparative advantage of the country in the availability of cheap and skilled labour force, to tap the export market. In this connection, we refer to Srinivasan’s (1998) observation that it is not possible to expect significant export spillovers from FDI in

Notes

1 For example, the peak customs duty rates were reduced from 150% in 1991-92 to 30.8% in 2002-03 (for more details see Ahluwalia 2002).

2 See Heckman (1981).

3 Published by Central Statistical Organisation (CSO), government of India. The Input-Output (IO) absorption matrix (1998-99) is used to compute the αjh for the first two periods (1993-96 and 1997-2000) u nder study and the IO absorption matrix (2003-04) is used to compute the αjh for the last two periods (2001-04 and 2005-08).

4 Independent samples T test in SPSS is used for this analysis.

5 In linear regression, an outlier is an observation with large residual. In other words, it is an observation whose dependent variable value is unusual given its values on the predictor variables (Chen et al 2006).

6 Leverage is a measure of how far an independent variable deviates from its mean (Chen et al 2006).

7 This is consistent with the findings of Joseph (2005).

References

Aggarwal, Aradhana (2002): “Liberalisation, Multinational Enterprises and Export Performance: Evidence from Indian Manufacturing”, The Journal of Development Studies, Vol 38(3), pp 119-37.

Ahluwalia, M S (2002): “Economic Reforms in Indian since 1991: Has Gradualism Worked?”, Journal of Economic Perspectives, Vol 16(3), pp 67-88.

Aitken, B J, G Hanson and A E Harrison (1997): “Spillovers, Foreign Investment and Export Behaviour”, Journal of International Economics, 43, pp 103-32.

Athreye, Suma and Sandeep Kapur (2001): “Private Foreign Investment in India: Pain or Panacea?”, The World Economy, Vol 24(3), pp 399-424.

Banga, R (2003): “The Differential Impact of Japanese and US Foreign Direct Investments on Exports of Indian Manufacturing”, ICRIER Working Paper No 106.

Caves, R E (1996): Multinational Enterprises and Economic Analysis, Surveys of Economic Literature (Cambridge: Cambridge University Press).

Chen, X, P B Ender, M Mitchell and C Wells (2006): Stata Web Books, UCLA Academic Technology Services, accessed from http://www.ats.ucla. edu/stat/stata/webbooks/, on 4 August 2008.

GoI (2008): Economic Survey 2007-08, Ministry of F inance, Government of India.

– (2009): Export Import Data Bank – TRADESTAT Version 6.0, Ministry of Commerce and Industry, Department of Commerce, accessed from http://commerce.nic.in/eidb/default.asp on 25 March 2009.

Greenaway, D, N Sousa and K Wakelin (2004): “Do Domestic Firms Learn to Export from Multinationals?”, European Journal of Political Economy, 20(4), pp 1027-43.

Economic & Political Weekly December 26, 2009

EPW

Heckman, J (1981): “Statistical Models for Discrete Panel Data” in C F Manski and D McFadden (eds.), Structural Analysis of Discrete Data with Econometric Applications, MIT Press.

Joseph, T J (2005): “Foreign Ownership and Export Spillovers in Indian Manufacturing Industry under Liberalisation”, chapter 10 in Allam Ahmed (ed.), the World Sustainable Development Outlook 2005, Global Competitiveness: A Common Goal in a Digital Society (UK: Emerald Group Publishing), pp 151-63.

Kokko, A, M Zejan and R Tansini (2001): “Trade R egimes and Spillover Effects of FDI: Evidence from Uruguay”, Weltwirtschaftliches Archiv, 137, pp 124-49.

Kumar, N and J P Pradhan (2003): “Export Competitiveness in the Knowledge-based Industries: A

Firm-level Analysis of Indian Manufacturing”, Discussion Paper, Research and Information System (RIS) for Developing Countries, New Delhi.

Kumar, N and N S Siddharthan (1994): “Technology, Firm Size and Export Behaviour in Developing Countries: The Case of Indian Enterprises”, Journal of Development Studies, 31(2), pp 289-309.

Montgomery, D C, E A Peck and G Vining (2004): Introduction to Linear Regression Analysis, Wiley Series in Probability and Statistics, John Wiley and Sons, Inc.

Pradhan, J P and V Abraham (2005): “Attract-

I ndia (like in east and south-east Asian countries), since India’s infrastructure sector is less efficient compared with many of those countries with whom India competes in international markets. Poor infrastructure like ports (both air and sea), road networks, etc, makes it less feasible to export, because such costs would cancel out the competitive advantage from the locationspecific factors such as cheap factors of production. However, r ecent policies like the special economic zone policy, and i ncreased investments in export-related infrastructure, are e xpected to attract more export-oriented FDI. This would not only help the foreign firms to set up their export base in the country, but also help the domestic firms to reduce their exporting costs and to become more competitive.

Appendix 1

Smarzynska, B K (2004): “Does Foreign Direct Investment I ncrease the Producti vity of Domestic Firms? In Search of Spillovers through Backward Linkages”, American Economic Review, Vol 94(3), pp 605-27.

Srinivasan, T N (1998): “India’s Export Performance: A Comparative Analysis” in I J Ahluwalia and I M D Little (ed.), India’s Economic Reforms and Development Essays for Manmohan Singh (Delhi: Oxford University Press).

Topalova, Petia (2004): “Trade Liberalisation and Firm Productivity: The Case of India”, IMF, Working Paper WP/04/28, International Monetary Fund.

UNCTAD (United Nations Conference on Trade and Development) (1999): World Investment Report 1999 (New York and Geneva: United Nations).

– (2002): World Investment Report 2002 (New York and Geneva: United Nations).

ing Export-Oriented FDI: Can India Win the Race?”,

Intensities of Domestic and Foreign Firms for the

Working Paper No 156, Gujarat Institute of Deve-

Period 1993-2008 by Industry Groups

lopment Research, Ahmedabad.

Industry Group Mean Difference t - Value

RBI (2007): Statistical Handbook, Reserve Bank of India, obtained from www.rbi.org.in on Food products -0.283 -0.256 10 April.

Textiles -16.648** -9.000

– (2009): Reserve Bank of India Bulletin, obtained from www.rbi.org.in on 30 March.

Chemicals 1.878* 2.574

Ruane, F and J Sutherland (2004): “Foreign Direct Non-metallic minerals -1.454** -3.476 Investment and Export Spillovers: How Do Export

Metal and metal products -4.842* -2.711

Platforms Fare?”, International Institute of Integration Studies (IIIS), Discussion Paper No 58,

Basic machinery -3.904** -6.314

Trinity College, Dublin. Electrical and electronics -4.082** -7.447 Sjoholm, Fredrik (1999): “Do Foreign Contacts Enable

Automobiles 0.082 0.212

Firms to Become Exporters?”, EIJS, Working Paper No 68, Stockholm School of Economics,

N 16

Sweden. Significant at** 1%, and* 5% level.

vol xliv no 52 105

Performance Variable Between Period 1 Between Period 2 Between Period 3 (1993-96) and Period 2 (1997-2000) and (2001-04) and (1997-2000) Period 3 (2001-04) Period 4 (2005-08)

F@ t /z# F@ t /z# F@ t /z#

Part A: Comparison of Performance of Domestic Firms in Different Time Periods under Study

Average export intensity 8.68** -2.28** 2.879a -1.338 2.59a -1.83a

Average R&D intensity 0.384 -0.532 3.70* -1.169 7.96** -1.74a

Average technology import intensity 10.98** 2.03* 37.06** 4.37** 86.12** -6.84**

Average value added per labour 7.66** 4.23** 18.69** 6.25** N A N A

Average value added per capital 16.04** 6.55** 0.142 2.53** N A N A

Part B: Comparison of Performance of Foreign Firms in Different Time Periods under Study

Average export intensity 0.517 -0.82 0.68 0.130 6.68** -2.01*

Average R&D intensity 5.2* -1.68 a 1.82 0.88 5.37* -1.173

Average technology import intensity 7.34** 1.65a 18.17** 3.46** 18.78** -3.63** Average value added per labour 21.44** 3.62** 2.34 2.41* N A N A

Average value added per capital 1.351 2.36* 1.26 1.83 N A N A

@ F test is used to test the equality of variances. # To test the equality of mean differences, t-statistic is used if there is no significant differences in the variances, otherwise standardised z-statistic is used.

Appendix 2: Testing for Difference in the Export

Dear reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here

Comments

(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top