post-autistic economics review
Issue no. 30, 21 March 2005
article 3



      issue 23 contents                            PAE Review index                               home page




Liberalisation and Social Structure:
The Case of Labour Intensive Export Growth in South Asia

Matthew McCartney   (SOAS, University of London)

© Copyright 2004 Matthew McCartney


Neo-classical theorists argue that discrimination is impossible in a competitive market economy.  Any firm or individual with a ‘taste for discrimination’ will be driven out of business by lower cost competitors who employ, trade and produce according to the criteria of profit and productivity maximisation.  By this logic, free markets and free trade will allow a developing country to exploit a comparative advantage in labour intensive manufacturing and agro-processing.  Such labour intensive growth will inevitably draw women for the first time into employment outside the home.  Employed women will achieve an independent income and higher social status.

Neoclassical economics forgets economics is a branch of social theory.  The outcome of liberalisation will be crucially dependent on the social structure of values and institutions in which it occurs.  Where there is a pre-existing ideology of gender subordination (India and Pakistan) business has utilised this social structure to cut costs and fragment formal labour institutions.  More egalitarian Sri Lanka has witnessed women being drawn into formal sector employment and receiving the benefits of independence, mobility and independent income.

Liberalisation, Efficient Growth and Labour Markets

In Neoclassical economics growth in a free market will reflect the preferences of rational individuals, growth must then by definition be efficient1, each exchange will reflect mutually beneficial gains by optimising economic agents. ‘Getting the prices right’ will allow developing countries to exploit a comparative advantage in labour-intensive production and exports.  Overvalued exchange rates discriminate against labour-intensive agro-exports and reduce the cost of imported capital goods.  Minimum wage and other labour ‘rights’ raise the cost of labour and reduce the elasticity of employment with respect to output growth.  Sometimes synonymous with removing ‘urban bias’, liberalisation could be better described as removing capital bias.

Numerous authors have highlighted labour market rigidities as being a key factor in explaining poor developmental outcomes in India2.  Employment protection3 has hindered restructuring in union dominated public sector enterprises.  It became more costly and time consuming for firms to adjust to changing market conditions and absorb new technologies, legislation encouraged firms to remain small and informal4, trading off access to formal credit and scale economies in order to avoid the strictures of labour legislation.  Firms in declining industries were prevented from shedding excess labour and piled up losses; those in expanding industries were reluctant to hire new workers and so substituted labour for capital.  The jobless growth in the 1980s is cited as evidence for this proposition.  Minimal expansion of formal sector employment between 1980 and 1989 coincided with an increase in annual earnings per worker of 3.5% and large-scale substitution of labour for capital5.

The supposed potential of labour-intensive growth in South Asia is demonstrated by post-Mao China.  China’s labour markets have proved highly flexible in the non-state sector.  Formal sector employment increased rapidly from 95m in 1978 (9.7% of the economically active population) to 148.5m in 1994 (19.2%).  In India by contrast formal sector employment has increased from only 22.9m in 1978 (6.8%) to 27.4m in 1994 (5.4%)6.

Labour Intensive Export Growth and Gender

The crucial ‘optimistic’ Neoclassical assumption is that markets reflect mutually beneficial voluntary exchange.  The spread of markets is thus by assumption proof that the economy is becoming more efficient.

Discrimination in the sense of paying workers of identical productivity different wages will not persist over the long term in a capitalist economy.  Any employer refusing to employ workers on the base of colour, creed, gender or caste will be less profitable than a non-discriminatory rival.  Over time the dynamics of competition will drive out those with a ‘taste for discrimination’, the logic of the market will separate discrimination from the process of production.  In a broader sense the dictates of the market separate society from the economy.

“The bourgeoisie, wherever it has got the upper hand, has put an end to all feudal, patriarchal, idyllic relations.  It has pitilessly torn asunder the motley feudal ties that bound man to his ‘natural superiors’, and has left remaining no other nexus between man and man than naked self-interest, than callous cash payment’7.

‘Female Employment Intensive’ Growth

Historically labour intensive export growth has been associated in many instances with a disproportionate increase in the employment of women.  In the textile mills of nineteenth century Britain, silk weaving in 1920s Japan, and electronics factories in the post-war Tiger economies.  Wood8 found a strong relationship between increased exports and increased female employment in manufacturing, the largest increases occurring in Mauritius, Tunisia, Sri Lanka, Malaysia and the East Asian Tiger Economies.  In Export Processing Zones (EPZs) most labour is female, eighty percent in the Caribbean and the Philippines; this bias is especially strong in the garments sector9.

Drawing women into the formal labour market will improve their economic status and social position.  The relative respect and regard of women is strongly influenced by their ability to earn an independent income, being employed outside the home and having ownership rights.  These factors are linked by the positive impact they have on strengthening women’s voice and agency through independence and empowerment.  With independent waged employment the contribution of the woman to the family’s well-being is more visible, she is less dependent on others, the exposure to ideas outside the home makes her agency more effective10.  An ability to seek employment outside the home can contribute to the reduction of women’s relative and absolute deprivation.

Really Existing Markets

Neoclassical economists are quick to generalise this optimistic scenario to support an exclusive focus on liberalisation.  However, re-rooting economics as social theory reveals such outcomes to be context dependent.  Markets as they really exist do not accord to this hypothesised ideal, rather, they are embedded in wider social structures of values and institutions.

“Real markets are permeated by power relations of various kinds; they are embedded in social processes which may, for example, involve class exploitation or gender subordination; and they are saturated by divergent institutions, ideologies, ethical and cultural values.11

The idea that liberalisation will remove politics from the economy and lead to a more rational and efficient allocation of resources is false.  Markets are not politically neutral but are embedded in social structures.  Just as government intervention can be distorted by an underlying political economy so too can markets. There is no neat dichotomy between state-regulation and market, rather both are meshed into existing social structures of (among others) caste, religion and gender12 in South Asia.

Liberalisation can remove market constraints but not structural constraints such as patriarchal values that prevent equal access of men and women to markets.  These are not just imperfections of the market but deep-rooted characteristics of society13.

Social Construction of Gender in India

In India employers have utilised a pre-existing and intensifying ideology of gender subordination to undermine male unionised labour and replace it with low cost ruralised, casualised, and informalised female labour.  In 1985, 250,000 people were employed in the Bombay textile mills, by 1996 this had declined to only 54,000.  Women then constituted only 0.01% of the cotton mill workforce while accounting for more than 45% of the unorganised cotton handloom sector14.

The social environment in which ‘labour intensive’ growth occurs is a crucial determinant of its net impact on women’s agency. The missing factor to be controlled for is the pre-existing social structure of gender relations.  The experiences of Sri Lanka and Pakistan offer sharply contrasting responses to liberalisation.

Sri Lanka and Pakistan: Empowerment and Exploitation

Sri Lanka has had traditionally good human development indicators, a 90% plus literacy rate and, compared with other developing countries, minimal gender disparity in education15.  There has been a persistent upward trend in the educational attainment of males and females. Female literacy in 1995 was 87%, only marginally lower than that of men.  In this context, of relatively equal gender relations liberalisation has had an empowering impact.

Liberalisation policies were initiated in 1977. By the early 1980s there was economic growth of slightly under 6% per annum.  The fastest growing sectors since 1977 have been unskilled labour intensive manufactures, mainly garments and textiles.  The structure of exports has indeed shifted towards Sri Lanka’s comparative advantage.  The institution behind this growth has been mushrooming Export Processing Zones (EPZ’s)16, set up in 1978, 1984 and 1990.  80% of the employment in EPZ’s is female.  The female-male ratio in manufacturing has increased from 25% to 80% between 1963 and 1985, total employment of women in manufacturing increased by 50% between 1977 and 199517.

In Pakistan by contrast, gender relations have historically been highly unequal, the sex ratio at 910 being even lower than that of India.  Literacy, school enrolment and a persistently high fertility rate all point to the low status of women.  Female primary school attendance is around 35%, among the lowest in the world.   Female labour force participation is only 3.5% and represents severe crowding into low pay, low-skill occupations18.  The labour market is also highly segmented, especially in urban areas where there is also widespread segregation between sexes.

The structure of modern industry in Pakistan is similar to that of Sri Lanka, consisting mainly of labour intensive processing of agricultural products as well as textiles and clothing.  In 1992/93 textiles and garments accounted for 64% of total export revenue19.  However on a pre-existing base of gender discrimination the gender composition of employment in textiles and garments is very different.  In Pakistan it is a male intensive sector; 88.3% of urban textile manufacturing workers are men. As in India there is evidence of widespread contracting by garment enterprises that employ poor and young women.  Most of this work is done in informal sector workshops or home based work.  Such work by itself will have little impact on bargaining strength or increasing the social status of women.  In fact there may be a contradictory impact by implicitly devaluing the implied worth of unpaid domestic employment and further marginalising those women without paid external employment.

“[I]n South Asian countries, women are rarely able or willing to work outside the marital home.  In that case, wage work outside the house increases her work load because she is not in a position to bargain with others about sharing her housework.20

Gender and Fragmentation of Labour in India

Structural adjustment in the 1990s has given impetus to the long-term trend of increasing market relations in South Asia.  There has been no long-term improvement in the status of women in Pakistan or India.  The logic of capital is not dissolving discrimination but working within the social structure of a pre-existing gender ideology and intensifying female disadvantage.

Female-Male (population) ratios have been declining in India since the beginning of this century.  The 1991 Census showed a further decline in this ratio to 927 women for every 1000 men21.  Mortality levels of women are abnormally high from birth until the mid-30s. Increased urbanisation, modernisation and economic growth have not improved these trends.  In fact the lowest ratios are recorded in the richest states, Haryana and the Punjab.  Using the ratio for Sub-Saharan Africa, Dreze and Sen22 calculated there were approximately 37 million missing women in India in 198623.

Economic change is working within a social structure of female disadvantage.  The north Indian pattern of anti-female discrimination is spreading southwards by means of cultural assimilation, and is not being undermined by economic change.  Even in North India there is a pronounced process of Sansrikisation24.  The dominant castes in the north are the martial and patriarchal Rajputs and Jats.  Traditionally they have a pronounced gender division and obsession with honour.  Honour is to a large extent a function of the conservative behaviour of women25.  Lower castes in the north have been traditionally more equal, female-male ratios among tribes and Scheduled castes having long been significantly higher.  Over time lower castes have been emulating, not opposing, the ideology of dominant castes.  Practices such as restrictions on widow re-marriage and dowry have been diffusing down the caste hierarchy.  The fall in female-male ratios has been generated by a convergence of ratios in lower castes to those of dominant castes26.


Liberalisation is context dependent, not a neutral and deterministic process.  The social structure within which liberalisation occurs has a crucial impact on outcomes.

Despite having levels of female disadvantage of a similar if not worse magnitude to India and Pakistan, throughout the 1990s Bangladesh has enjoyed rapid labour (female) intensive employment growth in the export-orientated textiles sector.  There is evidence to support the proposition that a prior restructuring of social relations (by NGOs among others) has enabled this favourable outcome27.

Economics is social theory.  Neoclassical theory forgets social structure at the cost of relevance.


For a longer exposition see Matthew McCartney, ‘Driving a car with no steering wheel and no road map: Neoclassical discourse and the case of India’, Post-Autistic Economics Review, 21, 13 September (2003) article 5,

2. See for example Peter R.Fallon and Robert Lucas, ‘Job Security Regulations and the Dynamic Demand for Labour in India and Zimbabwe’, Journal of Development Economics, 40 (1993), pp. 241-75; Roberto Zagha, ‘Labour and India’s Economic Reforms’, in J.D.Sachs, A.Varshney and N.Bajpai (eds) India in the Era of Economic Reforms (Oxford University Press, 1999), pp. 160-185; Montek S. Ahluwalia, ‘Economic Reforms in India Since 1991: Has Gradualism Worked?’, Journal of Economic Perspectives, Vol 16 No.3 (2002), pp. 67-88; Timothy Besley and Robin Burgess, ‘Can Labour Market Hinder Economic Performance? Evidence from India’, LSE, STICERD Working Paper No.33.

3. The 1976 amendment to the Industrial Disputes Act required state government permission to carry out any retrenchments in a firm of more than three hundred employees.  An amendment in 1982 reduced this figure to one hundred.  The constitutional validity of this change was challenged and the second reform reintroduced as a constitutional amendment in 1984, ZaghaLabour and India’s Economic Reforms’.

4. Only 7% of a labour force approaching 390m are in the organised sector subject to social security and labour laws.

5. R. Nagaraj ‘Organised Manufacturing Employment’, Economic and Political Weekly, Vol 35, No 38 (2000), pp. 3445-8.

6. Nirupam Bajpai, ‘Sustaining High Rates of Economic Growth in India’ Harvard Centre for International Development, Working Paper, No 65, March (2001).

7. Karl Marx and Friedrich Engels, ‘The Communist Manifesto’, (Penguin, 1967), pp82.

8.  Adrian Wood, ‘North-South Trade and Female Labour in Manufacturing: An Asymmetry’, Journal of Development Studies, Vol 27, No. 2 (1991).

9. Martin Rama ‘Globalisation and Workers in Developing Countries’, World Bank Development Research Group, Working Paper No. 2958 (2003), pp. 15.

10. Amartya .Sen, ‘Development as Freedom’, (Oxford University Press, 1999), Ch 8.

11. Gordon White, ‘The Political Analysis of Markets: Editorial Introduction’, Institute of Development Studies Bulletin, Vol 24, No 3 (1993) pp. 1.

12. Barbara Harriss-White, India Working: Essays on Society and Economy, (Cambridge University Press, 2003).

13. Indira Hirway , Economic Reforms and Women’s Work’, in Employment’ in T.S.Papola and A.N.SharmaGender and Employment in India’, (Vikas, 1999), pp 356.

14. Ushma Upadhyay ‘India’s New Economic Policy of 1991 and its Impact on Women’s Poverty and AIDS’ Vol 6, No. 3, Feminist Economics, (2000), pp.108.

15. Marzia Fontana, Susan Joekes and Rachel Masika, ‘Global Trade Expansion and Liberalisation: Gender Issues and Impacts’, Briefings on Development and Gender, Report No. 42, University of Sussex (1998), pp.23.

16. An EPZ is geographical location where foreign and/ or local investors are allowed to set up 100% export-orientated facilities, usually an array of incentives are granted such as tax holidays, access to duty free imported inputs and superior infrastructure.

17. Wood, ‘North-South Trade and Female Labour in Manufacturing: An Asymmetry’, pp. 168-89.

18 Shahid. J. Burki, ‘Pakistan: Fifty Years of Nationhood’, Third Edition, (Westview, 1999), Ch 4.

19. Fontana et al, ‘Global Trade Expansion and Liberalisation: Gender Issues and Impacts’, pp. 25.

20. Nirmala BanerjeeHow Real is the Bogey of Feminisation?’ in T.S.Papola and A.N.SharmaGender and Employment in India’, (Vikas, 1999), pp. 314.

21. Satish.B.Agnihotri, ‘Missing Females: A Disaggregated Analysis’, Economic and Political Weekly (1995), 19th Aug.

22. Jean Dreze and Amartya Sen, ‘India: Economic Development and Social Opportunity’, (Oxford University Press, 1995), pp. 141.

23. There are a number of explanations for these disturbing trends, see Harriss-White, ‘India Working: Essays on Society and Economy’.

24. A process whereby lower castes emulate the practise and rituals of higher castes.

25. See Jean Dreze and Haris Gazdar, ‘Uttar Pradesh: The Burden of Inertia’, in J,Dreze and A.Sen (eds), ‘Indian Development: Selected Regional Perspectives’, (Oxford University Press, 1996), pp. 33-128.

26. The female-male ratio among the dominant Rajput caste has remained stable at 890 for most of this century, there has been a general convergence among lower castes to this ‘standard’.

27. See Petra Dannecker, ‘Between Conformity and Resistance: Women Garment Workers in Bangladesh’, (The University Press Limited, 2002).

Matthew McCartney,
“Liberalisation and Social Structure: The Case of Labour Intensive Export Growth in South Asia”, post-autistic economics review, issue no. 23, 5 January 2004, article 3,