ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Girl Child Protection Scheme in Tamil Nadu: An Appraisal

The Girl Child Protection Scheme in Tamil Nadu was introduced in 1992; surveys and data analysis show that between the late 1990s and 2002-03, daughter elimination has declined sharply. However, a close look at the scheme reveals that its implementation is not targeted at districts with a high prevalence of female infanticide, that it assumes only poor families are anti-daughters, and given the sterilisation condition, that families with only daughters and strong son preference are not likely to volunteer. Also, to what extent it has altered attitudes towards daughters - one of its aims - is unclear.




’” -






organisations (NGOs) have been active in

Girl Child Protection Scheme

terms of data collection efforts to track gender differences in s urvival, and in in

in Tamil Nadu: An Appraisal

troducing programmes to prevent daughter elimination, some of which have been adopted in other states or have been intro-Sharada Srinivasan, Arjun S Bedi duced on an all-India basis. In particular,

The Girl Child Protection Scheme in Tamil Nadu was introduced in 1992; surveys and data analysis show that between the late 1990s and 2002-03, daughter elimination has declined sharply. However, a close look at the scheme reveals that its implementation is not targeted at districts with a high prevalence of female infanticide, that it assumes only poor families are anti-daughters, and given the sterilisation condition, that families with only daughters and strong son preference are not likely to volunteer. Also, to what extent it has altered attitudes towards daughters – one of its aims – is unclear.

Sharada Srinivasan ( is assistant professor, International Development Studies, York University, Canada. Arjun S Bedi ( is professor, Development Economics, Institute of Social Studies, Erasmus University Rotterdam, The Netherlands.

ver the years, the phenomenon of India’s “missing women”, a term used by Amartya Sen (1990), has attracted a lot of attention amongst policymakers, activists and the academic community. Controversy surrounds the estimates of the extent of missing women. Jha et al (2006) claimed that as many as half a million daughters were eliminated every year. While the exact figure continues to be debated, it is clear that since 1961, the proportion of girls to boys or the 0-6 sex ratio has declined from 976 to 927 in 2001. The sex ratio at birth has also declined sharply and estimates based on the N ational Family Health Surveys (NFHS) and Sample Registration Surveys (SRS) show that this ratio has fallen from 943 in the 1980s to 876 in 2001-03, reflecting continued movement away from the n ormal sex ratio at birth of 952 girls to 1,000 boys. A substantial proportion of the d ecline in these sex ratios may be a ttributed to the differential survival of girls and boys due to daughter elimination in the form of sex selection, infanticide and neglect.

While several north Indian states have a long history of daughter elimination, Tamil Nadu is a relatively recent entrant to the list of states exhibiting the phenomenon. Notwithstanding its recent a ddition, the state government and non- governmental

november 28, 2009

in 1992, following the continued efforts of NGOs and the media the Tamil Nadu government acknow ledged the prevalence of daughter elimination and announced s everal schemes to “eradicate” it. These included the Cradle Baby Scheme (CBS) which allows families to hand over unwanted f emale babies to the government, legal a ction against perpetrators of infanticide and the Girl Child Protection Scheme (GCPS) which provide financial incentives to families with only daughters. In a ddition to these state-wide interventions, in 1997-99, in Dharmapuri, a special, oneoff behavioural change campaign with k alaipayanams (itinerant street theatre) at its centrepiece was used to create awareness, to highlight the value of girls and to m obilise the population against female i nfanticide. In addition to government - led interventions, several NGO-led initiatives have been operating in the districts of M adurai (including Theni), Salem (including N amakkal) and Dharmapuri.

One Half of Humanity

This article provides a brief review of the patterns of daughter elimination in Tamil Nadu and then goes on to focus on the role played by, in particular, the government’s GCPS in shaping sex ratios and influencing daughter elimination. This is important from a policy perspective as the Tamil

vol xliv no 48

Economic & Political Weekly


Nadu scheme is the longest running scheme of its kind in the country; it is the forerunner of a number of schemes introduced in other states, for example, Delhi and Haryana’s Ladli Scheme, Madhya Pradesh’s Ladli Lakshmi Yojana, and A ndhra Pradesh’s GCPS. In a national meeting on “Save the Girl Child” on 28 April 2008, the prime minister noted that, “No nation, no society, no community can hold its head high and claim to be part of the civilised world if it condones the p ractice of discriminating against one half of humanity represented by women.”

( 677

a ccessed on 30 Octo ber 2009.) He went on to add that the various schemes undertaken by certain state governments, such as the “Dikri Bachao campaign of Gujarat, GCPS of Tamil Nadu, Devi Rupak Scheme of Haryana, Ladli campaign of Delhi and the scheme for cash incentives to panchayats for improving the village sex ratio of Punjab are good steps”. Notwithstanding the remark that these schemes are “good steps”, little is known about the effectiveness of such schemes. Similarly, while much has been written about the resources allocated to Tamil Nadu’s GCPS, including in the Hindu (24 January 2009), whether the scheme has achieved its

o bjectives is not known.

In Tamil Nadu, daughter elimination in the form of female infanticide first came to public attention in the mid-1980s. In 1997, Chunkath and Athreya estimated that there were about 3,000 deaths per year due to social causes. Initially it was felt that the practice was limited to certain groups and certain geographical areas but subsequently several researchers showed that the practice was widespread. Based on our analysis of various rounds of Vital Events Surveys – these surveys are unique to the state and provide information on male and female live births and infant deaths based on a sample of about 9 million individuals – we find that b etween 1996 and 2003, the incidence of daughter elimination has declined sharply from a deficit of about 4,485 girls every year between 1996 and 1999 to about 2,000 in 2003, a decline of 46% between the two periods. The declines are dominated by Dharmapuri and Salem, the two districts which account for the highest proportions

Economic & Political Weekly

november 28, 2009

of daughter deficit (Bedi and Srinivasan 2008). Other sources such as the Sample Registration Surveys and the N ational Family Health Survey also suggest a decline in daughter deficit and a recent report (2005) submitted to the State Planning Commission based on primary health centre (PHC) records also note that the number of female infant deaths due to “social causes” declined from an average of about 3,000 a year between 1995 and 1999 to 372 in 2002, that is, a d ecline of about 88%. While the exact magnitude of the decline may be debated, it is clear that between the late 1990s and the early years of the current decade there has been a sharp d ecline in the extent of daughter elimination. Recent rounds of field work condu cted by the authors in 2007 in Salem and neighbouring districts confirm the durability of the reduction in daughter elimination.

While the pattern of reduction in daughter elimination is clear, the role played by various interventions in bringing about the decline in daughter elimination has not been intensively examined. Previously, in 2000, Athreya and Chunkath analysed the effect of the behavioural change campaign in Dharmapuri and concluded that the campaign had a sharp effect on reducing infanticide. According to the figures provided by the social welfare department, between 1992 and November 2007, about 2,400 girls have been surrendered to the CBS (Srinivasan and Bedi 2009). The scheme has served as a safety valve for unwanted girl children. However, the role played by the GCPS, which aims to a ddress deep-rooted concerns like attitudinal change towards daughters has not been analysed. We now turn to this scheme.

Missing the Target?

The GCPS was launched in 1992. According to government documents the aim of the scheme is to promote family planning, eradicate female infanticide and promote the welfare and status of girl children in poor families. It is based on the assumption that given the perception of girls as an economic burden, it is necessary to e nhance their economic value by providing financial support to families that bring up daughters. It is targeted at families below the poverty line, who have only

vol xliv no 48

daughters in the age group of 0-4, and no sons, and if either of the parents has undergone sterilisation before the age of 35 years. The government deposits Rs 2,000 on behalf of eligible families in an interest-bearing fund. Money from this fund was to be paid out to families on

o ccasions such as the first birthday of the child, on joining school and on joining class VI with a terminal payment of Rs 10,000 at the age of 20. A sum of Rs 40 million per year was allocated for the scheme. During 1992-97, the programme uptake was limited; only 2,039 families had benefited from the scheme.

In 2001-02 the GCPS was restructured to confer increased financial benefits. Currently, in the case of poor families with only one daughter and no son, and where either parent has undergone sterilisation before the age of 35 years, the government deposits Rs 22,200 and in the families with two girls and no son, Rs 15,200 each, for 20 years in the Tamil Nadu Power F inance and Infrastructure Development Corporation. Interest accruing from this deposit provides monthly payments of about Rs 150 to the family and a terminal payment, at age 20, of Rs 80,000 for a one-girl family and a benefit of Rs 40,000 per girl for families with two girls. Based on income limits for scheme eligibility, the sum of Rs 150 per month leads minimally to a 15% increase in income for families with two girls and a 4% increase for families with one girl. A sum of Rs 227 million, in real terms 2.83 times more than the a llocation in 1991-92, was allocated for the scheme in 2001-02. The scheme has witnessed a 50-fold increase in the number of beneficiaries and according to the corporation’s policy note 2006-07, a sum of about Rs 1,750 million benefiting 1,15,171 children had been received under the scheme between 2001 and mid-2006.

While the number of beneficiaries has increased sharply, the effect of such a p rogramme on altering attitudes towards girls and reducing daughter elimination is unclear. Available district-specific data (Table 1, p 12) show that the programme had beneficiaries from all the state’s districts and in 2001-02 and 2002-03, only 15% of the beneficiaries were from the five high infanticide districts of Dharmapuri, Madurai, Salem, Theni and Namakkal,


while 22% of the beneficiaries were from seven districts (Chennai, Coimbatore, Kancheepuram, Kanyakumari, Nilgiris, Thanja vur, Thiruvarur) where there is limited (or no) evidence of the practice. More formally, there is a very weak correlation (0.13) between the extent of daughter deficit in 1996-99 and the number of GCPS beneficiaries, suggesting that the programme was not targeted towards districts with a high prevalence of daughter elimination.

The principle of providing economic support to protect girls may be appropriate but the implementation of the scheme and the eligibility criteria suggest that the GCPS may not be achieving its objectives. First, as discussed above, the scheme is not targeted at districts with high prevalence of daughter elimination. Second, the GCPS targets families below the poverty line and frames daughter elimination as a problem amongst the poor. However, existing evidence shows that daughter deficit is not more likely to occur in poorer families and indeed may be more prevalent in better-off families. Based on regression analysis of village-level data from Salem district, we show that the probability of daughter elimination is not affected by i ncome, wealth, education and caste. In a 2003 analysis based on NSS data Siddhanta et al (2003) show that at the national level and in many large states the f emale to male 0-14 sex ratio declines with increasing levels of expenditure. Data from Census 2001, analysed by Bhat and Zavier (2007), reveal that the sex ratio at birth is more likely to be masculine for mothers with more education. Education and prosperity appear to facilitate knowledge of, access to, and the use of technology for sex selection. Third, although participation in the programme is voluntary, given the sterilisation clause, families with a strong son preference and who have only daughters are unlikely to be attracted to the scheme. The possibility that the scheme attracts those who do not have a strong son preference is consistent with the district-wise distribution of the larger share of GCPS beneficiaries in districts with limited evidence of daughter deficit. Notwithstanding these shortcomings, similar schemes have been introduced by several state governments. While such cash transfers may increase the ability of poor families to educate and take care of their girls and raise the value of daughters, whether it has indeed done so is an unanswered question. However, it is unlikely that a scheme such as the GCPS which focuses only on the poor – while daughter elimination is more widespread

– and imposes sterilisation as an eligibility requirement will have a substantial effect on reducing daughter elimination.

Reconsideration Needed

If the scheme is to specifically address bias against daughters that leads to their elimination, then the challenge is how to increase uptake in districts where daughter aversion is acute. To ensure greater uptake and prevent daughter elimination, the sterilisation and the “no son in the family” clauses will have to be reconsidered as these may actually work against daughters, forcing families to choose between the

Table 1: GCPS Beneficiaries and Daughter Deficit in Tamil Nadu

Districts Distribution of Distribution of
Beneficiaries (%) Daughter Deficit (%)
2001-03 1996-99
Chennai 3.06 1.16
Coimbatore 5.45 0.49
Cuddalore 5.97 1.8
Dharmapuri 2.83 18.13
Dindigul 3.25 5.86
Erode 3.76 4.37
Kancheepuram 4.35 1.4
Kanyakumari 5.38 0.71
Karur 3.71 4.29
Madurai 2.50 7.83
Nagapattinam 3.64 3.22
Namakkal 2.32 4.69
Nilgiris 0.86 0.07
Perambalur 1.97 1.24
Pudukottai 2.01 1.2
Ramanathapuram 3.54 0.54
Salem 4.46 25.62
Sivagangai 0.72 1.81
Thanjavur 1.95 2.09
Theni 2.67 3.51
Tirunelveli 4.98 6.53
Tiruvarur 1.59 1.85
Thiruvannamalai 4.80 1.02
Thiruvallur 4.91 0
Tiruchirapalli 2.01 0.22
Tuticorin 3.02 2.08
Vellore 4.46 3.68
Virudhunagar 2.71 1.05
Villupuram 6.98 6.94

Source: Directorate of Social Welfare, ROC No 58544/CW6/ 2007, dated 3 January 2008 and authors’ calculations based on Vital Event Surveys.

november 28, 2009

scheme (daughters) or sons. Although Tamil Nadu has been the trendsetter in terms of schemes to protect girls, other states have pushed the envelope further. For example, in March 2007, the government of Madhya Pradesh launched its GCPS (Ladli Lakshmi Yojana) which was targeted at poor families and included sterilisation as a condition for eligibility. However, in April 2008, the government dropped the sterilisation condition and has now extended the scheme to all nonincome taxpayees. The result was a sharp uptake in number of beneficiaries. Whether relaxation of these conditions has worked towards reducing daughter elimination still needs to be examined.

While the decline in daughter elimination experienced in Tamil Nadu is laudable, the discussion here shows that given the targeting mechanism and the sterilisation condition linked to GCPS, it is unlikely that the scheme has played a substantial role in reducing daughter elimination, and it is also unlikely that without rethinking some of the scheme’s conditions that it can be expected to play a large role in reducing daughter elimination in the future.


Bhat, Mari and A J Francis Zavier (2007): “Factors I nfluencing the Use of Prenatal Diagnostic Techniques and the Sex Ratio at Birth in India”, Economic & Political Weekly, 42(23), pp 2292-2303.

Bedi, Arjun Singh and Sharada Srinivasan (2008): “Tackling Daughter Deficits in Tamil Nadu”, India. Available at

Chunkath, Sheela Rani and V B Athreya (2000): “Tackling Female Infanticide: Social Mobilisation in Dharmapuri, 1997-99”, Economic & Political Weekly, 36(49), pp 4345-48.

Chunkath, Sheela Rani and V B Athreya (1997): “Female Infanticide in Tamil Nadu – Some E vidence”, Economic & Political Weekly, 32(17), WS-21-8.

Jha, Prabhat, Rajesh Kumar, Priya Vasa, Neeraj Dhingra, Deva Thiruchelvam and Rahim Moineddin (2006): “Low Male-to-Female Sex Ratio of Children Born in India: National Survey of 1.1 Million Households”, Lancet, 9 January, DoI: 10.1016/S0140-6736(06) 67930-0, available at

Sen, Amartya (1990): “More Than 100 Million Women Are Missing”, The New York Review of Books, 20 December.

Siddhanta, S, D Nandy and S B Agnihotri (2003): “Sex Ratios and ‘Prosperity Effect’: What Do NSSO Data Reveal”? Economic & Political Weekly, 38(41), pp 4381-4404.

Srinivasan, Sharada and Arjun S Bedi (2009): “Ensuring Daughter Survival in Tamil Nadu, India”, available at wp 477.pdf. (The Hague: Institute of Social Studies).

State Planning Commission (2005): “Round Table Discussion on Human Development and Agriculture Diversification and Water Resources Management, 18-19, May (mimeo) (Chennai: State Planning Commission), workshop/6HD%20and%20Health-SEC.PPT, accessed on 30 October 2007.

vol xliv no 48

Economic & Political Weekly

Dear Reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top