Why reservation is necessary
  Sukhadeo Thorat

back to issue

WITH a rapid scaling down of our tiny public sector due to privatisation and increasing withdrawal of the state under the impact of liberalisation, serious concern has been expressed about the fate of the present public sector reservation policy. Because of indirect and backdoor de-reservation, there is a growing demand for some sort of affirmative action policy for the private sector which so far has remained outside the purview of any kind of anti-discriminatory measure and in which more than 90% of dalit and adivasi workforce is engaged. The issue found a place in the election manifesto of political parties and in the Common Minimum Programme of the present government. Drawing from theoretical and empirical literature on the issue of economic discrimination, I will try to provide reasons for a reservation policy for the private sector as a remedy against discrimination in labour, capital and other markets and indicate general guidelines for framing such a policy.

Discrimination on the basis of race, religion, ethnicity, national or social origin exists in many nations under diverse social, economic and political systems. In order to correct the imbalance in terms of access to capital assets, employment, education, political participation and other spheres, countries have turned to practices of reservation, affirmative action, positive action or equal opportunity policies for these discriminated sub-groups in addition to general pro-poor policies. A great majority of these policies and programmes of intervention operate in respect to sub-group populations identified by ethnic, racial, religious or gender characteristics. The examples are not only from the West (USA, UK, Northern Ireland and Yugoslavia) but also from Latin American countries like Brazil, Bolivia, Peru; African countries like Nigeria, Sudan, South Africa and countries like Malaysia, Pakistan, China, Japan and India from Asia. It is surprising that while the affirmative action policy in many of these countries was, to begin with, used for both private and public sector, the India state never thought of bringing the private sector under the purview of a reservation policy – even though it is the fact of discrimination in the private domain that led us to accept the reservation policy for the public sector.

Why are governments in developing and developed countries concerned about exclusion and discrimination? Why do they develop policies against such practices? Is discrimination only an issue of equity or does it also involve economic costs to the society? Are the costs it imposes on the society more social and political than economic? The insights from mainstream economic theory indicate that economic, particularly market, discrimination has multiple consequences; it hampers economic growth, induces income inequality and creates potential for inter-group conflict by denying equal opportunity to discriminated groups (Birdsall and Sabot 1991). Therefore, remedies against discrimination – legal, affirmative action or compensatory in nature are required both for equity and economic growth. I will argue that we need a reservation policy for private sector in India for reasons of equity as well as economic growth.



Let us first discuss the case for reservation from the point of view of economic growth. All standard theories of economic discrimination predict adverse consequences of market discrimination on economic growth through their detrimental consequences on profits, wages and efficiency in the allocation of labour. They suggest that economic discrimination will slow down growth by reducing efficiency due to sub-optimal allocation of labour among firms and economy, by reducing job commitment and effort of workers who perceive themselves to be victims of discrimination and, by reducing the magnitude of investment in human capital by discriminated groups, returns on this investment. (Birdsall and Sabot 1991).

The consequences of discrimination on inequality are far more evident and therefore justification for reservation/affirmative action policy from equity consideration is clear. Denial of access to resources, employment, education and common facilities that others have, it impoverishes the lives of individuals from excluded groups and is a clear denial of certain basic human rights. The concern about discrimination thus is an issue not only for equity but also for economic growth.



Ambedkar pioneered the reservation policy as early as the 1920s, mainly on the basis of widespread and ubiquitous discriminatory and exclusionary character of Hindu society with devastating consequences, particularly for former untouchables. The caste system’s characteristic feature of fixed and compulsory occupation (or property rights) with concomitant fixed economic rights for each caste, implies ‘exclusion’ of one caste from undertaking the occupations of other castes. Exclusion and discrimination is in fact internal to the economics and sociology of the caste system and its natural outcome. In terms of consequences, the economic theories of the caste system clearly predict negative outcomes of caste and untouchability based market discrimination for economic growth and income distribution (Akerlof 1976, Scoville 1984, Lal 1991, and Ambedkar 1987).

Fixed occupations essentially involve restrictions on mobility of labour and capital across caste groups, leading to an imperfect market situation and a fragmentation of economic activities. Akerlof-Scoville-Lal’s theoretical model thus predicts that given the segmented and imperfect character of the labour market, the economic outcome of the caste economy is lower than posited in the model of perfectly competitive markets.

Ambedkar added that efficiency and productivity of labour is adversely affected by multiple disincentives involved in customary rules of castes. The economic pursuit in a caste system is not based on individual choice, sentiment and preference, in so far as it involves an attempt to allocate a task to an individual in advance, selected not on the basis of training or capacity but on the social status of parents. The social and individual efficiency requires us to develop the capacity of an individual to the point of competency to choose and make one’s own career. This is missing in the overall scheme of the caste system.

Further, some of these occupations are considered polluting or impure and therefore socially degrading. The social stigma of impurity and pollution reduces the social status of persons engaged in them and thus lowers the economic incentives. ‘What efficiency can there be in a system under which neither men’s hearts nor their minds are in their work?’ (Ambedkar, 1936). As an economic organisation caste is therefore a harmful institution, in as much as it involves the subordination of man’s natural powers and inclination to the exigencies of social rules.



The consequences are particularly pronounced in terms of income distribution, employment and poverty experienced by the excluded/discriminated groups. Since property rights under the caste system are assigned unequally across castes, income distribution is generally skewed along caste lines. The unequal and hierarchical assignment of occupation and property rights among castes implies that although every caste, except those at the top of the caste order, suffers to various degrees from an unequal division of social and economic rights, the former untouchables, who are located at the bottom of the caste hierarchy, suffer most as they face ‘exclusion and discrimination’ from access to all economic rights, including right to property, except manual labour or service to the castes above them.



Beside the general negative impact on income distribution, labour immobility across occupation also adversely impacts employment. Ambedkar (and Akerlof in a different context, 1980) argued that by restricting mobility of labour across caste occupation and thereby not permitting readjustment of employment, caste becomes a direct cause of ‘voluntary unemployment’ among high caste persons and ‘involuntary unemployment’ among the low caste persons. A high caste Hindu would generally prefer to be voluntarily unemployed for some time than to take up an occupation not assigned to his caste. On the other hand, for low caste untouchables the restrictions to take other caste’s occupation compels them to remain involuntarily unemployed.

Thus, judged by the standard criterion of economic efficiency, the caste system as an economic organisation lacks all elements required to fulfil the conditions for optimum economic outcome. The caste and untouchability based economic discrimination have serious consequences on economic development, income distribution, right to individual development and equal right to employment, all of which cumulatively have poverty-inducing consequences, particularly for the low caste untouchable.



Reducing economic discrimination thus becomes essential because it is likely to increase economic growth, provide equal access to discriminated groups, reduce inequality between groups and minimise the potential for conflict which inequality between groups may give rise to. What are the remedies against market discrimination? Conclusions regarding the consequences of market discrimination on economic growth and income distribution are derived from mainstream economic theory. The same theory also predicts that in highly competitive markets, discrimination will prove to be a transitory phenomenon as there are costs associated with discrimination to the firm/employer which result in lowering profits. Firms\employers who discriminate, face the ultimate sanction imposed by the markets. This theoretical perspective thus posits the resultant erosion of profits as a self-correcting dimension of discrimination.

The free market solution is not, however, a practical remedy as, discrimination might persist, particularly in the labour market, over long periods with or without prevalence of a free market situation. First, not all markets are highly competitive. The persistence over decades of labour market discrimination in high income countries attests to that. Indeed, in developing countries, employers have significant monopoly power to discriminate at will. Second, even if competition exists in all markets, is not a sufficient condition for the elimination of discrimination if all employers are discriminators.



These two theoretical viewpoints have obvious policy implications. Those who believe that discrimination is indeed self-correcting argue for strengthening competitive market mechanisms. But if discrimination continues to persist despite competitive market process (which in reality is the case) or for other reasons mentioned above, market interventionist policies will be necessary.

‘The analytical stance of the mainstream neoclassical economists is characterized as methodological individualism and it presumes that economic institutions are structured such that society-wide outcomes result from an aggregation of individual behaviours. It presumed that if individuals act on the basis of pecuniary self-interest then market dynamics dictate equal treatment for equal individuals regardless of inscriptive characteristics such as race. Consequently, observed group inequality is attributed to familial, educational, or other background differences among individuals who are unevenly distributed between social groups. The causes of a dissimilar distribution of individuals between social groups may be genetic, cultural, historical, or some combination thereof. The differences in cultural attributes include the value families and neighbourhoods place on education, attitudes, and work habits. The historical refers primarily to the impact of past discrimination on current inequality.

‘In contrast, economists who may be classified as methodological structuralists do not accept this interpretation. Structuralism as an analytical method holds that aggregate outcomes are not the result of a simple summation of individual behaviours, but rather arise from the constraints and incentives imposed by organizational and social hierarchies. In this view, individual behaviour achieves its importance within the context of group formation, cooperation, and conflict. Economic and political outcomes are thus a function of the hegemony exercised by dominant groups, the resistance offered by subordinate groups, and the institutions that mediate their relationship. Discrimination, in this view, is an inherent feature of the economic system. Competition is either not powerful enough to offset the group dynamics of identity and interest, or it actually operates so as to sustain discriminatory behaviour. Discrimination is due to the dynamics of group identification, competition, and conflict rather than irrational, individual attitudes. Market mechanisms, far from being relied upon to eliminate discrimination of their own accord, must be scrutinized and pressured to further the goal of equality of opportunity’ (Shulmen and Darily 1989).

The policy implications of this view on persistence of discrimination are obvious. Since the markets will continue to operate in an imperfect manner, discrimination will persist. It will also persist even if market forces are competitive in nature under certain conditions mentioned above as a normal phenomena; it therefore calls for intervention in the form of an affirmative action policy and other measures as safeguards against discrimination.



The customary regulatory framework of the institution of the caste system and untouchability, on which mainstream theoretical formulations are based, has now undergone significant change. After the adoption of the Constitution in 1950, in theory at least caste-based customary rules and norms governing occupation, property rights, employment and wages, and education were formally replaced by an egalitarian legal framework of property rights under which the ‘low castes’ now have equal access to all occupations, education and other spheres. However, despite these provisions and the impact of other factors, the caste system continues to function in the private domain of economy in modified and changing forms and therefore safeguards are required against market and non-market discrimination.



The corporate sector, however, has by and large opposed affirmative action of any sort on grounds that it does not discriminate in hiring practices. It has further argued that a reservation policy of the type used in the public sector in India will reduce competitiveness and efficiency. It is clear from the earlier discussion that this argument is neither based on economic theory nor on empirical evidence about the working of labour and other markets. While the corporate sector advances the agenda of liberalisation based on support from mainstream economic theories, it refuses to accept lessons from the economic theory of discrimination. In particular it refuses to recognize the need for market intervention in the form of affirmative action to overcome market imperfections caused by caste based discrimination and to induce market competitiveness.

Empirical studies on the working of labour and other markets, and social needs like education, housing and health services, provide evidence of the persistence of market discrimination, particularly of former untouchables, and its end result in the form of lack of access to fixed capital assets, employment, human development and culmination in high poverty and deprivation among them. (Action Aid study 2005, Thorat 1999, reports of the SC/ST Commission). The studies also bring out the exclusionary and discriminatory working of private industrial labour markets (Papola 2004).

So the claim of the corporate sector that it follows fair and competitive hiring practices is not borne out by facts. More than 60% of recruitment in organised industrial sector is through informal modes of recruitment which are exclusionary in their outcome. Affirmative action is thus necessary for promoting competitiveness and economic growth, if not for the goal of equal opportunity. An efficiency conscious corporate sector cannot avoid the lessons from the theories which they use as justification for pushing the liberalisation agenda.



How and in what forms we need to extend the existing public sector reservation policy for discriminated groups in the private sector is, however, as crucial. There is the huge experience of our own public sector reservation policy. The Indian state could also learn from the measures used against discrimination in private sector of other countries in the world. Drawing on both Indian and international experience, I wish to spell out the possible elements of reservation or affirmative action policy for the private sector in India.

In designing remedies against discrimination we need to clearly distinguish between legal, equal access and other positive measures. These aspects relate to (a) the type of economic sector or market for which the legal and affirmative action measures are developed and (b) the type of method or procedures used in their application for the private and public sector.

First, with respect to the economic spheres or markets, countries such as USA, Northern Ireland, South Africa and Malaysia have mainly developed policies for religious, racial and ethnic minority groups. Broadly speaking, in these countries multiple economic and social spheres are covered under the orbit of legal and affirmative action measures which include labour, agricultural land, capital, product and consumer goods markets and also the transactions in supply of social goods, such as education, housing and the transactions undertaken by the government with private minority businesses. The specific economic spheres covered vary from country to country.



In countries like USA and Northern Ireland, where the non-agriculture sector constitutes more than 90% of the workforce, the focus is mainly on affirmative action policies for the labour market, both in public and private sectors. In some countries like USA, besides the labour market, legal and affirmative action measures also cover education, housing, and government contracts for construction and purchase of goods from minority businesses. In developing countries like Malaysia and South Africa where a substantial portion of population is engaged in the agriculture sector, and minority groups suffer from poor access to land and capital in addition to the labour market, the affirmative action measures are also extended to agriculture land and capital market in addition to affirmative action policies for basic social needs like education and housing.

A second aspect relates to methods and procedures used to operationalise the safeguards against discrimination. Countries have used at least three kinds of procedures or methods. First is the legal protection against discrimination in the form of Equal Employment Opportunity Laws (EEOL).



These laws prohibit any private or public employers from discriminating against workers or persons based on group identities like religion, gender, colour, ethnicity, national and social origin and provide legal safeguards to discriminated groups in the event of discrimination in employment and other spheres of economic activities. Article VII of the Civil Act which established Equal Employment Opportunity as law (EEOL) (Executive Order 11246) in USA, and Fair Employment Act in Northern Ireland, are some examples.

A second strategy relates to affirmative and positive actions of various type. In principle, affirmative action can be distinguished from other anti-discrimination measures requiring proactive steps to ensure fair participation of discriminated groups (in employment and other spheres like education, housing, government contracts etc), in contrast to laws that only prevent employers from taking steps that disadvantage minorities in the labour market and other spheres. Given the limitations of EEOL in ensuring fair participation of discriminated groups, they are supplemented by affirmation action and positive measures, which attempt to ensure a fair share to discriminated groups.

The principles (and methods) applied to judge ‘fair or just participation’ in employment, educational admission, political participation or government contract and other spheres vary among the countries. Generally speaking, the population share of minority groups constitutes the basis to judge fair participation or access. In some cases just participation is viewed in term of fixed quotas (similar to India), in other cases it is expressed in term of racial/religious minority (population or labour force) balance, and ‘appropriate candidate pool’ with numerical goals and timetables without quotas. Further, these fixed goals or targets are made legally mandatory or compulsory in some cases while in others they are pursued with an element of voluntary action on the part of firms. In both cases, however, some sort of enforcement machinery is designated to monitor goals and targets. The office of Federal Contact Compliance in USA and FAIR Employment Agency in Northern Ireland are some examples.

A third strategy is ‘reparation or compensation’. Reparations and compensation are defined as payment for an acknowledged grievous social injustice to a group (Darity 1997). It is necessary to recognize that different affirmative action policies are generally deployed as a measure and safeguard against ‘present discrimination’. This has its limitations in overcoming the effects of discrimination and exclusion suffered by a community in the past for long periods of time. The Equal Opportunity Act and affirmative action programmes of various kinds which intend to provide legal protection and ensure just participation in the present are inadequate to compensate for wrongs done in the past, resulting in complicit resourcelessness. Therefore, some countries have used selective compensation as a method to pay and empower excluded and discriminated groups as a one time settlement.



An overview of the strategies against economic and social discrimination used by different countries indicates three remedies – namely equal (employment) opportunity laws, reservation/affirmative action measures and reparation and compensation, either together or in combination. Adoption of some or all of these remedies against discrimination inflicted on low caste untouchables and other groups in the private sector in India will be conditioned by the nature of economic and social discrimination.



In my view the reservation policy for the private sector, namely agriculture, private industry and service sector, and cooperative sector where more then 90% of SC and ST are engaged should by guided by three principles. It should be applicable to multiple spheres, fixed quotas with some kind of monitoring mechanism and, depending on the nature of discrimination, using all three instruments – legal, fair access and compensation – in combination.

It is necessary that the government enact an equal opportunity law to provide legal safeguards against discrimination for various markets, namely capital, agriculture land, employment, product and consumer goods, education, housing, health and others. The legal safeguards should then be supplemented by more positive and reservation/affirmative actions to improve their effective access to private employment, agricultural land, capital, production and consumer goods, and private education.

Since these reservation/affirmative action policies address the issue of present discrimination, they do not generally help to compensate for historical exclusion. This can be addressed within the framework of reparation or compensation measures. The former untouchables are a potential case for reparation or compensation due to denial of property rights and other rights and injustice for long periods in history, which is reflected in their overwhelming presence as wage labour, insignificant share in business and low literacy and educational levels.



George Akerlof (1976), ‘The Economic of Caste and of Rat Race and Other Woeful Tales’, Quarterly Journal of Economics, XC.4, November 1976.

George Akerlof (1980), ‘The Theory of Social Customs, of Which Unemployment May Be One Consequence’, Quarterly Journal of Economics, XCIV.4, June 1980.

B.R. Ambedkar (first published 1987), ‘Philosophy of Hinduism’, ‘The Hindu Social Order – Its Essential Features’, The Hindu Social Order – Its Unique Features’, Vasant Moon (ed.) Dr. Babasaheb Ambedkar Writings and Speeches, Vol.3, pp. 1-94.

Biswjit Banerjee and J.B. Knight (1985), ‘Caste Discrimination in the Indian Urban Labour Market’, Journal of Developing Economics, ?????.

Faridah Jamaluding, ‘A Study of the Malaysian Economy: The New Economic Policy (1971-2000) and National Development Policy (1991-2000)’, Journal of Social Science and History, Fall 1988.

Deepak Lal (1984), Hindu Equilibrium: Cultural Stability and Economic Stagnation, Vol. I, Clarendon Press, Oxford.

Nancy Birdsall and Richard Sabot (1991), ‘Unfair Advantage – Labour Market Discrimination in Developing Countries’, World Bank Studies, ???????.

T.S. Papola ( 2004), ‘Social Exclusion and Hiring Practices by Private Industrial Sector’, Paper presented a Seminar on Remedies Against Discrimination in the Context of Reservation in the Private Sector, IIDS, Delhi, August 2004.

James G.L. Scoville (1996), ‘Labour Market Underpinnings of a Caste Economy – Failing the Caste Theorem’, The American Journal of Economics and Sociology 55(4), October 1996.

Amartya Sen (2000), ‘Social Exclusion: Concept, Application, and Scrutiny’, ADB Working Paper.

S.K. Thorat (1999), ‘Caste and Labour Market Discrimination’ (with R.S. Deshpande), Indian Journal of Labour Economics, Conference Issue, November.

S.K. Thorat (1996), ‘Ambedkar on Economics of Hindu Social Order: Understanding Its Orthodoxy and Legacy’ in Walter Fernandes (ed.), The Emerging Dalit Identity, Indian Social Institute, Delhi.

William Darity Jr. and Steven Shulman (1989), Question of Discrimination – Racial Inequality in the U.S. Labour Market, Wesleyan University Press, Middletown, Connecticut.