A partnership with academia

Building knowledge for trade and development

Vi Digital Library - Text Preview

Does Openness Promote Competition—A Case Study of Indian Manufacturing

Discussion paper by Manoj Pant and Manoranjan Pattanayak, 2005

Download original document (English)

Technical paper on a test of the hypothesis that openness by itself promotes competition.

Does Openness Promote Competition—A Case Study of Indian Manufacturing

Manoj Pant and Manoranjan Pattanayak *


July, 2005

* Professor of Economics and Ph.D. student, respectively

CITD, School of International Studies

JNU, New Delhi –110067

E-mail: manojp@mail.jnu.ac.in


This paper uses firm level data for the period 1989-2001 to analyse the working of competition in India’s manufacturing sector. Some simple econometric models are used to see the impact of greater competition on profit markup over the last decade. The econometric analysis of the factors determining markup indicates that, contrary to received wisdom trade openness by itself does not act to reduce profit markup. This may, however, reflect the fact that the exporting firms being more efficient are able to sustain larger markups. In other words, the degree of efficiency is itself the entry barrier. Moreover, if anything, it is the presence of foreign firms in an industry that acts in a competitive way to reduce markups. The paper also investigates the degree of competitiveness defined as the Lerner price-cost margin. The analysis indicates that the estimated margins are in general high over the nineties across all industries and in most of the industries considered these margins have been increasing over the second half of the 1990s. In addition, the standard entry barriers like returns to scale do not seem to be operative.

Our analysis thus indicates that the market by itself does not bring about competitive outcomes probably because of the unequal strengths of the market players. This indicates that the regulatory agencies probably have a crucial role to ensure a level playing field.

Does Openness Promote Competition—A Case Study of Indian Manufacturing

1 Introduction

An important feature of developing country policies in the last decade has been enactment of Competition Acts as part of a pro-competition policy (see, Hoekman, 2002). While a multilateral competition policy is no longer a part of the agenda of the WTO, the Doha Declaration of 2001 relates competition policy to trade policy and emphasizes reinforcement of competition policy in developing countries through capacity building.

India too has substantially improved the competition climate in its manufacturing sector since 1991 via a series of changes in both domestic and trade polices. However, as avenues for competition have increased at the same time the government has also put in place a set of regulatory agencies with the objective of promoting competition. As trade theorists are prone to argue, if the economy is open to market forces, then the market players would themselves enforce competitive outcomes. This would be particularly true for countries ( like India) which are classified as small countries in the world economy with a share of world trade under 1 percent. If this proposition is accepted then it would follow that the only objective of regulatory agencies would be setting the rules of the game rather than monitoring the degree of competition. Yet, the terms of reference of the regulatory agencies often include a clause requiring the active promotion and monitoring of competitive forces ( see, for example, Competition Act, 2002; Electricity Act, 2001).

One hypothesis commonly quoted in the literature is that imports can raise the level of competition in the domestic industries and reduce profit mark-up. The theoretical framework and econometric specification for this kind of study are provided by Levinsohn (1993) and Metin-Oxcan (2000). However, the empirical evidence on this is at best inconclusive.

Levinsohn in his paper tests the imports-as-market discipline hypothesis in case of Turkey in the light of reduced import protection in the decade of the 1980s and reduction in tariffs, though to a lesser extent in some of the manufacturing industries. The estimation of an oligopolistic structural equation model reveals price mark-up (=price/marginal cost) has declined in imperfectly competitive industries in which protection has declined and increased in imperfectly competitive industries in which protection has increased. This result Levinsohn takes to conclude that the results can be taken as evidence for supporting imports as market discipline hypothesis.

In a similar exercise, Krishna & Mitra (1998) try to study the impact of trade liberalization on certain sectors of the Indian manufacturing sector, namely, Electronics, Electrical Machinery, Non-electrical Machinery and Transport Equipment. The estimated markups (defined as Profit/ (Raw Material Cost+ Energy Consumption Expenditure+ Wages) (data for firm taken from CMIE) decline after liberalization except in electrical machinery industry thereby confirming the market discipline hypothesis. However, the effect of liberalization on technology and hence productivity is ambiguous depending upon the conditions of the home country. However, there also exists some literature that does not support the hypothesis very clearly.

Jayanthakumaran (1999) in a study pertaining to Australia runs separate regressions to find the effect of liberalisation on labor productivity growth, export growth and changes in price-cost margin1. He found that labour productivity is negatively, but weakly related to change in effective rate of protection. The growth in exports has been negatively related to effective rate of protection and positively related to internal demand variable (implying improvement in productivity and increased intra industry trade) and index of scale variable (implies increased impact of entry barriers). However, the result that stands out is that the influence of change in effective rate of protection on price cost margins has been negative implying that openness has not reduced the price cost margins. For this two explanations have been forwarded by Jayanthakumaran. Firstly, it could be a case where there is a monopoly element at work in the Australian industry. Moreover, the effective protection estimate is the mere reflection of the pricing policy, which in this case would tend to diminish slowly overtime. Secondly, the effect of labour productivity on price cost margins is positive and hence the author concludes that the increase in labour productivity has been transformed into higher price cost margins.

Harrison (1994) has tested for changes in productivity and market structure due to trade liberalisation in the manufacturing sector of Cote de’Ivoire. In a Cournot setting of firm behaviour she found that markups (price/ marginal cost) do decline in the heavily protected sectors after trade liberalising measures but not in the case of other sectors. She also found that price cost margins are higher in sectors with lower import penetration and higher tariffs. Changes in productivity growth due to openness were also assessed using three approaches, viz., differentiating between pre reform and post reform, low and high tariff regimes and low and high import penetration period. It was found that open trade policies increased productivity.

Metin-Ozcan et al (2000) also analysed the changes in the market concentration and accumulation patterns in case of Turkey for the period 1980-96. They found that openness has a very small, but negative influence on industrial concentration. They also found that openness reduces the profit margins, though to a lesser extent, while concentration and real wage costs increases the profit margins to a greater extent. The study concludes that the outward orientation of the economy has not brought much benefit in terms of the extent of reduction in industrial concentration and the authors attribute this to the institutional and technological barriers in the country.

Roberts (1998) has dealt with the changes in industrial concentration in Poland between 1988-93 to evaluate the effects of trade liberalising measures in 1989. She found that though the average concentration for the industry as a whole has declined in the period, in some sectors the concentration has actually increased. The sectors in which concentration has increased were previously less concentrated, with high capital- output ratio and shrinking demand. They also found that among the less concentrated industries, competition has increased in the consumer industries. The latter trend is due to tougher price competition.

The objective of this study is three fold. First to quantify how policy changes have increased the competiveness of the Indian manufacturing sector. Second, to estimate the role of trade policy in promoting competition and, third, to estimate the changes in the degree of monopoly power in Indian manufacturing industry in the last decade. In the following section we indicate our main data base. In section III we first investigate if the changes in domestic and trade polices since 1991 have in fact promoted competition. In Section IV we outline the explanatory model used to estimate the impact of polices on a measure of competition. Here we also asset out and estimate a model to measure the degree of monopoly power. Finally, in Section V some policy conclusions are noted.

II. Database

The primary source of data for the empirical exercises of this paper is the PROWESS database of the CMIE (Centre for Monitoring the Indian Economy). This database contains information on about 8000 companies, which includes companies that are public, private, co-operative, joint stock, listed or otherwise. This wide coverage encompasses almost seventy percentage of the economic activity of the organised sector both manufacturing and non-manufacturing (our focus remains essentially on the manufacturing sector of the Indian industry). CMIEs’ methodological framework for data standardisation, via formal validation and quality control render inter-year, inter-industry and inter-company data comparable. The variables used in the equations of the models below have been constructed from data taken from the PROWESS database. The econometric testing of Section IV are based on balanced panels. The time period of our study is 1989-2001 for which data was available.

Section III. Policy Changes after 1991.

The most important policy change after 1991 was the Industrial Licensing Policy of 1991 which significantly improved the conditions of entry for both domestic and foreign firms. Our data base allows us to quantify this entry in terms of the date of incorporation of new firms after 1989. While this is an imperfect measure in that it does not imply new production starts or control for bogus firms it does indicate the intentions and perceptions of new firms. The data ( available till 2003) is given below.

Table 1

Gross Entry of New Firms in Manufacturing in India








Food and Beverages









Non-Mettalic Mineral Products



Metals and Metal Products






Transport Equipment






Source: PROWESS data base.

The data of Table 1 clearly indicates the big rush for netry of new firms particularly in the first period after 1989 till 1995. The reduced numbers after 1995 probably reflect the adjustment of new entry to the size of the domestic market and the fact that the growth rate of sales across all industries declined perceptibly after 1995. ( see, Das and Pant, 2004).

Another possible indicator of the changing nature of competition is the relative share of public sector enterprises (PSE) and private sector enterprises (PVT) in total sales. This is shown in table 2 below.


Share of Public Sector Enterprises in Aggregate Sales of the Industry: 1988/89-2000/01


April 1988 to March 1989

April 1994 to March 1995

April 2000 to March 2001