1.1 Background of the Indian Economy
Over the last two decades India has made a name for itself in the world economy as an emerging superpower. The inflexion point to this transformation began following the BOP crisis of 1991, when then Finance Minister, Dr. Manmohan Singh, introduced the policies of Privatization, Globalization and Liberalization to open up the Indian economy to the world. India’s openness ratio has skyrocketed since then with the majority of this openness stemming from a high energy import bill. Despite this, for its integration in merchandise trade, McKinsey Global Institute’s Connectedness Index ranks India behind emerging market peers such as Saudi Arabia, China and Russia. Importantly, the aforementioned peer economies are connected through exports, unlike India. Given India’s endowment of natural resources, India’s import bill may justify itself in the face of growing demands of an emerging economy (Shah, 2013). But the inelastic nature of Indian imports (oil, gas, precious metals, Fig A.1) further propounds the need for export performance to counteract the effects of high imports on the current account. Unfortunately, this has not been the case, and the economy has run a trade deficit since 1977 (Reserve Bank of India Statistics). Veermani and Gupta (2014) argue that keeping export growth at par with import growth is the only sustainable solution to the (trade) deficit problem. Therefore, even though exports as a share of GDP has grown reasonably, there exists tremendous requirement for export expansion. Whether this requirement is backed by potential to expand exports, forms the theme of this discourse. In light of ballooning imports, the merchandise trade balance deteriorated from -$2 billion to -$129 billion between 2001-2013 (RBI Statistics), escalating the need to diagnose the caliber of the export profile. While service exports have outperformed powerhouses like China and Japan, it still accounts for merely a dime out of every dollar exported by India. For the purpose of this paper, the focus would rest exclusively on merchandise exports over services given the relative importance of the former over the latter in overall trade.
The inability to expand exports despite evident potential to do so, deems India’s export performance to be sub-optimal. Such performance prompted Prime Minister Narendra Modi to launch “Make in India” in 2014, an export oriented strategy, incentivizing investments in the manufacturing sector to boost exports. The initiative can be viewed as being directed towards diversification of the Indian export basket to more sophisticated commodity exports. As part of the initiative, the Prime Minister also directed policy action to “(not just) look east (but also) link west”, which refers to strengthening ties with fast growing South Asian economies as well as Middle East and West Asia. Such policy action, targeted at stimulating more sophisticated commodity exports, as well as reorienting the geographic destination of these exports is resonant of where the inadequacies lay in India’s export setup. The analysis undertaken through this discourse, back the hypothesis that poor geographic orientation and relative under sophistication of the export basket are responsible for landing Indian exports in their current predicament. The subsequent recommendations formulated on the foundation of these analysis seek to serve as curative measures to resolve the economy’s export (and thereby deficit) conundrum.
1.2 Purpose and Contribution of this Paper
The aim of this paper is two fold. Firstly, to empirically analyse the anatomy of Indian exports through economic tools and expose any inadequacies in terms of overall quantum of exports, export destinations and export commodities. The paper opens by analyzing the key domestic and external factors affecting export propensity (exports to GDP). Fixed effects estimation helps furnish the standing of export propensity overtime. Fundamental to structural transformation is not just the question of overall quantum of exports but also geographic orientation and product composition of an economy’s exports (Anand, 2015). The analysis on geographical orientation evaluates import growth of India’s export partners, and adapts Hummels and Klenow’s (2005) intensive and extensive margins exports to geographies. The analysis on product space, rests on the results reflected by the Export Potential Index, which combines revealed comparative advantage and relative importance in world trade for each product exported by India. For the purpose of establishing a collectively exhaustive perspective towards Indian exports, I have sought to make my analysis in both absolute and relative terms. Relative comparison of exports, especially against emerging market economies institutes an informative yardstick on export performance. Secondly, I seek to leverage implications drawn from the aforementioned analysis to provide recommendations to improve export performance in the medium term.
1.3 Overview of the Paper
The rest of the paper commences by reviewing recent literature (Section 2) that dissects the anatomy India’s exports, through tools similar to the ones leveraged in this paper. Section 3 discuses the methodology, data and estimations used. Section 4 discusses the results along with implications derived. Section 5 concludes the paper.
2. Literature Review
Given that this paper covers analysis along three dimensions of Indian exports, the literature underpinning the premise is dispersed. While there is a variety of literature on determinants of export propensity, academic writings with respect to India’s export conundrum are relatively new. The launch of the “Make in India” initiative has channelized much needed focus on the importance of export enhancement. The synthesis drawn from the following empirical work, has implications which can be leveraged to narrow the trade deficit, and hence is critical to this paper. The sub-sections nested in this chapter first review most relevant literature on determinants of export propensity of a country, followed by recent relevant writings on Indian exports analysed along geographic and product dimensions.
2.1 Determinants of Export Propensity
Majeed and Ahmed’s (2006) empirical work on constructing estimates of export determination for 75 developing countries comes closest in terms of approach, to the model in this paper. They use a panel dataset from 1990-2005 and estimate the model using fixed effects. They seek to evaluate the importance of a range of domestic and external factors, on export propensity. For domestic factors, they find a positive and significant impact (at 5% level) for GDP, GDP growth, labor force, value added, official development assistance and proxies for development (number of people with access to TV, telephones). For external factors they found significant evidence that appreciation of REER adversely effects exports, and FDI inflows to have a positive effect (but not at statistically significant levels). The draw back of these models is that they don’t provide country specific estimates (although they do generate country specific intercepts), unlike simultaneous equation models.
Other empirical work on determinants employs the use of simultaneous equations models based on singular country specific factors. The shortcomings on these models is their dependence of relatively smaller datasets, which make estimates unreliable especially in light of poor reporting standards for statistics in certain developing countries. These models are also of little use for obtaining fitted values for other countries to make commentary on relative performance, given the characteristics of the countries. As a trade off simultaneous models provide specific country pertinent estimates, which is are useful in their own right. The estimates from some of the empirical work based on these models have broader implications, such as measures to narrow the trade deficit (Shah, 2013) and are thus relevant to this discourse. For instance, Sharma (2001) implements one such model for India with data from 1970-98 to deduce employable results indicating adverse effect of REER. He finds evidence for export demand decreasing upon export price rise relative to world prices. He also finds evidence that higher domestic demand reduces export supplies, but does not find statistically significant estimates for FDI inflows, though the sign is positive. Shah (2013) too uses a very similar model to that of Sharma (2001), but for a more recent time period from 1980-2011. He implements a 2SLS imperfect substitutes demand-supply simultaneous equations model. On the demand side he obtains unbiased results by instrumenting price of exports (proxied by export unit value index) by domestic prices (proxied by WPI). On the supply side Shah (2013) instruments price of exports by foreign prices (using export weighted REER index for India). Like Sharma (2001) he finds a significant negative relationship between price of exports and exports demanded. Shah’s (2013) findings indicate a 0.26% decline in exports demanded for a percent rise in export price. Furthermore, on the demand side his estimates indicate the demand for Indian exports to be elastic, rising 1.8% for every percent increase in world demand (proxied using import volumes weighted by partner share). In light of the importance of world demand, he recommends establishing closer ties to faster growing countries such as the Middle East and South Asia. On the supply side Shah finds strong evidence of elasticity between price of exports and supply of exports. His results also indicate a strong negative relationship between domestic prices rise and exports supplied, wherein a 1 percent rise in the former lead to a 3.1% decline in the former. In light of the last result, Shah (2013) points out the importance of the containing inflation/ domestic price level within the country. While several others conducted similar empirical studies leveraging simultaneous models for determination of export demand and supply for a country, the results obtained by Sharma (2001) and Shah (2013), are most pertinent to the cause of this paper.
2.2 Indian exports across geographies and product space
Anand, Kochhar and Mishra (2015), herein referred to as AKM, in a recent paper expressed the importance of India’s export diversification across commodities and geographies to structural transformation and export growth. The premise of their paper, is similar to this one. While the authors believe Indian exports have grown substantially, they find that there exists tremendous room for improvement through focus on the right commodities.
Their analyses document the evolution the composition of the export basket and diversification and relatedness of products for India’s export growth. The focus of AKM’s research lies on commodity analysis categorized by sophistication and technological content. They claim that sophisticated products not only contribute to economic growth through new products and destinations (extensive margin), but are fundamental to increasing the existing products in existing destinations (intensive margins). The authors analyse diversification of Indian exports along countries and commodities, given the potential of the medium to expand export growth. Based on the evidence that countries with exports concentrated in a few products, find it more difficult to expand and develop their revealed comparative advantage than countries that have more diversified exports (Hidalgo et al, 2007), AKM analyse relatedness of India’s exports.
The essence of AKM’s paper and my paper is similar, in that both believe exports to have grown substantially in the post liberalization period, but present tremendous room and requirement for development given the current predicament of the Indian economy. AKM’s research delves deep into analyzing the relatedness of products, relative to this paper which is less in-depth, but broader in its scope. AKM analyse evolution of product exports through analysis of quality and complexity in addition to the aforementioned dimensions.
Bhattacharya, Bruce and Mukherjee (2014), assert the need for rebalancing the economy, by bringing about a reversal to the “below-par performance” of late with respect to manufactured exports. They support the main motivation behind the “Make In India” program to invite FDI for export expansion. Concurrently, they stress the importance of keeping cost competitiveness while propelling the Indian manufacturing sector through FDI in light of forecasts indicating intense competition from emerging economies in the near future. Their analysis recommends the need to improve institutions, ease of doing business and government policies (lower FDI restrictions in certain sectors) in order to receive more FDI to foster manufacturing led transformation of exports. The authors hint at a domino effect, wherein better institutions (for instance), would lead to more FDI, which would lead to systematic transfer of knowledge leading to higher technology exports.
Veermani and Gupta (2014) study the role of intensive and extensive margins in market penetration for China and India between 1995-2011. For calculations along IM and EM, the authors adopt the approach developed by Hummels and Klenow (2005). The gist of their paper lies in furnishing an analysis to assert whether export promotion strategies should be directed towards increasing trade of existing relationships, or starting new trade relationships. They conduct an in-depth analysis of exports for the two countries by decomposing the intensive margin into price and quantity margins. They also assess export growth along the two margins by further disaggregating export profiles by broad economic categories and factor intensities. Their results indicate relatively better growth in the breadth of Indian exports (EM), which grew at 5% per annum, compared to 2% for the depth of exports (IM) between 1995-2011.They also find high penetration rates for natural resource industries. They find relatively low penetration for technology intensive exports, indicating an avenue for improvement. On the basis of the performance along the IM and EM, Veermani and Gupta recommend expanding exports along the intensive margin. This recommendation is contrary to the one in this discourse, mainly due to the approach adopted in analyzing Indian exports. However, both papers share the same school of thought when advising on bettering technology intensive exports.
3. Methodology, Data and Estimation
The essence of this discourse lies in bringing together methodologies that would be conducive to facilitate a discussion on export transformation. The methodologies are used to understand what affects overall export propensity of a country, favourability of geographic orientation of exports and potential of commodities exported. The results and implications derived from these analyses are used to analyse the underlying cause of India’s export expansion dilemma.
3.1 The Propensity to Export
Exports to GDP (herein referred to as export propensity) ratio is an informative starting point metric, to understand the role of exports for any given economy. More importantly, given that export promotion induces optimal allocation of resources (Majeed, Ahmed, 2006) it escalates the importance of understanding the levers would effect its movement. These levers can form the basis for formation of policy to promote exports. I implement the use of a panel set of countries encompassing majority of the countries in the world. I assess the movement of export propensity from fitted values (econometric specification to follow) between 1990-2014. For the purpose of these calculations, data was consolidated on exports, GDP, GDP per capita, population and real effective exchange rate from the World Bank database (World Development Indicators) and for inward foreign direct investment from UNCTAD database. The data is collected from 1990-2014, for 93 countries in order to make relative comparisons. The dataset was downsized from 214 countries by discounting countries with scanty data. The countries in the representative sample set encompass 83% of all world exports between 1990-2014, indicating reasonably small export shares of economies excluded from the dataset.
Leveraging the entire panel set of 93 countries from 1990-2014, helps gauge the impact of key internal characteristics as GDP per capita, population and external fundamentals such as real effective exchange rate and foreign direct investment inflows, on exports. The motivation behind this model to estimate whether India’s export propensity been in tandem with the economy’s fundamentals. The expectation is to find actual propensity to be above fitted values in recent years, given incremental policy action to promote exports over the last two decades. The use of a large set of countries is befitting, given it helps establish a better overall picture of what impacts an economy’s propensity to export on average, and to what magnitude. Moreover it enables us to assess actual versus fitted export propensity of peer economies and compare India’s performance to them. There exists a possibility of country specific invariant and unobserved characteristics affecting the explanatory variable. I largely mitigate this risk by estimating the model using fixed effects. I run the following specification
where, GDP PC is GDP Per Capita in current PPP dollars, population is the population of the country in a given year, FDI is inflows of foreign direct investment in a given year and REER is the real effective exchange rate index (2010=100). Ln indicates the variables in their natural logarithmic form
3.2 The Geographical Orientation of Indian Exports
The purpose of this dimension is to identify coherent relationship patterns between India and its export destinations. The question of whom a country exports is intertwined with how much a country exports as a whole. For instance, the overall quantum of exports will be higher if trade relationships are stronger with destinations with higher import demand and growth (Bachetta, 2012). Assessing the relationship between export share of each destination in India’s exports for 2014, with import growth for these destinations between 2005-2014, forms the first part of the analysis. Plotting the import growth rate of destinations over a decade long period, against their share in India’s exports in 2014 highlights what Lukauskas, et. al. (2013) call “demand pull effects”. The goal is determining whether exports are currently oriented towards high (positive geographic orientation) or low import growth economies (negative geographic orientation). In order to analyse share of export destinations in India’s total exports, and import growth of these destinations bilateral trade flows from WITS (World Bank) database. Using this data, year on year import growth was calculated for each year between 2005-14 and averaged out. Thereafter (log) average import growth of destinations (between 2005-14) was plotted against (log) export share of destinations to ascertain favourability of geographic exports.
Serving as an extension to this, the second part of the analysis involves adapting Hummel and Klenow’s (2005) approach to geographies to evaluate India’s exports along the (static) intensive margin and (static) extensive margins. This approach is different to Veermani and Gupta (2013), in that it is oriented towards margin (IM, EM) analysis for geographies rather than products. The intensive margin (defined as India’s export share in destination countries where it exports), is a helpful tool to not only assess expansion within existing markets over a time period, but also to make cross country comparisons with emerging market peers. For the purpose of calculating intensive and margins, the following formula by Bachetta (2012) was used (with data from WITS (World Bank)).
♣ IM : , where the numerator is the India’s exports its set of S destinations ), and the denominator is World Exports to the set S destinations (i.e India’s export partners). The IM helps understand India’s export share in countries where it exports (Hummels and Klenow, 2005).
♣ XM: : , where the numerator is World exports to India’s export destinations (S), and the denominator is total world exports (is the set of all countries in the world). XM is effectively the share of India’s destination markets in world trade (Hummels and Klenow, 2005).
3.3 Commodity Space Orientation and Composition of Indian Exports
The intent behind conducting an analysis of disaggregated commodity exports is to establish the current status and relative importance of Indian exports in the world market. This section embeds methodology for ‘product-destination’ orientation analysis, followed by assessment of existing commodities using the Export Potential Index.
In symmetry with the analysis on geographical orientation, a similar approach as used by Bachetta (2012) has been adopted to analyse product- destination orientation of Indian exports. The approach is additive to the one on geographical orientation in that it combines the latter with commodity exports. The share of each export commodity (say commodity ‘s’) to each destination (say destination ‘k) is plotted against the growth of world exports of each commodity category (say commodity ‘s’) to each destination (say ‘k’). Calculation are made on each product-destination pairing, for all commodities exported by India, to all its destinations. The export share of each product to a destination is calculated for 2014, along with corresponding import growth of a product by a destination over the last decade (2005-14) and transformed into their natural logarithmic form. The data for commodity used is ISIC2 at 3-digit level, from WITS (World Bank database). Import growth for each commodity to a destination was calculated year on year between 2005 and 2014, and aggregated to get average import growth rate of commodities to destinations between 2005-14.
The questions of how much a country exports and whom it exports to are greatly determined by what a country exports. This in turn effects the quantum of exports demanded, differing by geography. Product space analysis is especially critical for India, in that it it possesses tremendous room for development thereby effecting both the quantum and geographic orientation of its exports.
The second analysis leverages a methodology from the International Trade Centre (ITC) to obtain the Export Potential Index (EPI) based on Revealed Comparative Advantage. For the purpose of this discourse, 5000+ commodities were analysed at 6-digit level between 2008-14. The index focuses on analyzing existing products with most potential to increase export revenue. It can be thought of as serving to answer the question of which products to promote, to expand exports along the (product) intensive margin. Computation of 2 indices namely, Export Performance and World Demand together form the Export Potential Index with the following methodology.
The essence of the EPI is captured with the help of the above tree. Starting with a bottom up approach, the first step involves calculating the revealed comparative advantage for each product ‘p’ for India, transforming the RCAs into logarithm forms and then creating an RCA Index (normalised using X- Xmin/ Xmax –Xmin). Moving to the bucket number 2, a similar approach is adopted, wherein growth in RCA is calculated (between the 2008-13) and an index created for Growth in RCA. The weighted average of the two indices (RCA Index and Growth in RCA Index), form the Export Performance Index. Using exactly the same approach for buckets 3 and 4, Share (of a product in World exports) index and Growth of Share Index is created. The weighted average of the two forms the World Demand Index. The Export Potential Index is simply the average the of the export performance and world demand indices. The Export Performance Index, combines the RCA for a given product as well as growth in RCA over the span of period in question in order to rank the analysed product set in terms of their export performance. The world demand index, combines the share and the growth of share of the product before ranking them in order of their importance in the world. The results of this analysis are illustrated at chosen level of aggregation (6 digit), and industry classification.
...(download the rest of the essay above)