Foreign Involvement and Firm Productivity: An Analysis for Indian Manufacturing, Service, Construction and Mining Sectors

In recent years, the intensification in the global engagement of Indian firms, by way of exports and outward foreign direct investment (OFDI) has generated significant research that draws on theoretical propositions of the new new trade theory (Helpman, Elhanan, Marc J. Melitz, and Stephen R. Yeaple (2004), “Export versus FDI with Heterogeneous Firms”, American Economic Review, 94:1, 300–316.) (below HMY) that assign a leading role to heterogeneity in firm productivity in explaining self-selection of firms into foreign markets. In HMY, for the export versus OFDI decision, firms with the highest productivity are posited to invest abroad (Goldar, Bishwanath (2016), “Direction of Outward FDI of Indian Manufacturing Firms: Influence of Technology and Firm Productivity”, in Globalization of Indian Industries, Productivity, Exports and Investment, eds. De Beule, F., and K. Narayanan, India Studies in Business and Economics, 71–96, Springer.). Head and Ries (2003) (Head, Keith and John Ries (2003), “Heterogeneity and the FDI versus Export Decision of Japanese Manufacturers”, Journal of The Japanese and International Economies, 17:4, 448–467.) (below HR) note that an empirical complementarity between exports and OFDI could result with differences in fixed costs across destinations. Further, while for manufacturing, predictors such as physical transport cost and sunk cost of OFDI (as in HMY) are considered to be fairly standard, for services, different predictors, based on different considerations such as the need for direct communication with consumers, the difficulty of contracting nonroutine activities to foreign affiliates, near-zero transportation costs and non-commoditized products are proposed to reverse the HMY predictions (Bhattacharya, Rudrani, Ila Patnaik, and Ajay Shah (2012), “Exports versus FDI in Services”, The World Economy, 35:1, 61–78.). This paper examines whether involvement in OFDI is associated with higher productivity levels at the firm level (that is, whether OFDI firms are more productive than firms with purely domestic operations, and those that organize international activities only through exports). Cross-sectional findings of a positive link between firm productivity and foreign involvement could, however, be due to the most productive firms self-selecting into foreign markets, and/or learning effects through foreign engagements. Using Prowess database, over 1995–2010, in addition to manufacturing (57,698 observations) and services (5,145 observations), the under-investigated construction (2,036 observations) and mining (1,196 observations) sector firms are also considered, with no size thresholds. The non-parametric approach of first-order stochastic dominance (Kolmogorov–Smirnov test) is used to examine the nature of productivity differentials between firm categories (based on foreign involvement). In examining the main issue, this paper also discusses some measurement and/or methodological issues. Some modifications are applied towards the construction of real output (gross output (GO), value-added (VA)), and inputs series (combined intermediate inputs, namely, raw materials, energy and services, labour and capital) required for estimating total factor productivity (TFP). While following the widely used Annual Survey of Industries (ASI)-based approach to impute firm-level employment, an attempt is made to overcome the uniform wage criticism by adjusting the labour measure for a ‘wage premium’ based on ownership groups. The measure of physical capital allows for disaggregated growth of investment, and the capital stock measure combines physical and ‘knowledge’ or R&D ‘capital’ stock. In the context of productivity measurement, comparisons are drawn between the alternative methods that attempt to overcome ‘transmission bias’, namely, Levinsohn and Petrin (2003) (Levinsohn, James, and Amil Petrin (2003), “Estimating Production Functions Using Inputs to Control for Unobservables”, Review of Economic Studies, 70, 317–342.) (below LP) and its modification proposed by Wooldridge (2009) (Wooldridge, Jeffrey M. (2009), “On Estimating Firm-Level Production Functions Using Proxy Variables to Control for Unobservables”, Economics Letters, 104:3, 112–114.) (below WLP). Also, in the context of studies that point out that the relative superiority of exporters in comparison to purely domestic firms may also result from several sources of potential bias in productivity estimates (related to the selection of the functional form of the production function, namely, GO vs.VA), attempts are made to explore whether similar concerns are of importance when investigating the relative superiority of OFDI firms (that also export). Next, in the absence of information in the investment outside India data field in Prowess about the percentage holding by Indian firms in their affiliates abroad, while some studies identify an OFDI firm on the basis of existence of positive overseas assets, some use cut-offs on the fraction of OFDI to total assets (as for instance, >1%). In making the cross-sectional comparisons of the estimated productivity distributions, an attempt is made to see whether the stricter basis for classifying foreign investors affects the nature of productivity rankings by firm categories. For firms in the manufacturing and construction sectors, cross-sectional differences in TFP between outward investors that also export, pure exporters, and domestic firms are found to follow the HMY/HR hypotheses, although in contrast to the GO specification, the VA specification suggests an upward bias in the productivity advantage of internationally engaged firms (suggesting that controlling the ‘value-added bias’ is important and it is not sufficient to control only for the ‘transmission bias’). Productivity differentials vary, sometimes considerably by two-digit industry/industry groups. The HMY (and HR) pattern obtains, more so in textiles, coke and refined petroleum products, chemicals, pharmaceuticals, basic metal and fabricated metal, and machinery and equipment n.e.c. than in the rest. In services, TFP comparisons show that pure export firms dominate the purely domestic firms and overseas investors that also export dominate purely domestic firms. However, between the overseas investors that also export and pure exporters, no clear-cut differences could be established unlike a previous study for Indian software services suggesting the stochastic dominance of pure exporters over overseas investors that also export. This suggests that Indian IT firms’ OFDI that is mainly located in developed countries could also be guided by vertical or complex integration strategies, related to the technology-seeking motives and agglomeration economies (due to clustering in specific regions). In mining, only the dominance of pure export firms over purely domestic firms could be established for the latter half of the sample period. Qualified support is thus found for the ‘pecking order’ as predicted by heterogeneous firms’ theories. As the productivity and other firm characteristics of OFDI firms that initially start small are observed to be similar to those with larger positions abroad, if a constraint on financing is found to be an issue for these firms, the government should support a more liberal financial system for OFDI that could also aim specifically at firms with initially small OFDI. EXIM Bank (2017) (Export Import Bank of India (2017), “The Internationalisation of Indian Firms Through Outbound Foreign Direct Investment: Nature, Determinants and Developmental Consequences”, Occasional Paper No. 183. https://www.Eximbankindia.In/Assets/Dynamic/Pdf/Publication-Resources/Researchpapers/Hindi/82file.pdf .), for instance, indicates that there is a range over which it is possible to increase firms’ OFDI intensity and increase the benefits from OFDI.