INTERVIEW
‘We present a snapshot'
T.K. RAJALAKSHMI
Interview with Dr C. Chandramouli, Registrar General of India and Census Commissioner.
IT was with eager anticipation that the results of the 15th provisional national Census (the 15th since 1872) were awaited. Census 2011 has thrown up some interesting and also worrying results: literacy rates have gone up; fertility levels have declined; the child sex ratio (CSR) has worsened since the last Census, with several districts reporting regressive trends; the growth rate of the population in the Empowered Action Group (EAG) States has registered a significant decline; and the population of children (0-6 years) has declined. While other sociocultural and demographic data, including data on housing and household amenities, are yet to come, the Census office launched its caste Census in June following a government decision taken last year. Census 2011 also laid the foundation for the National Population Register (NPR). The caste enumeration exercise is expected to be over by the end of the year though it may take another six months for the findings to get tabulated and collated before being released.
Dr C. Chandramouli, the Registrar General of India and Census Commissioner, expressed concern over the declining CSR and spoke about the challenges of caste enumeration, among other things, in an interview to Frontline. Excerpts:
The provisional Census has described the decline in the CSR as an all-time historical low. The Census results provide policymakers with a direction to undertake follow-up action. While the improvement in literacy rates and the decline in fertility levels were widely welcomed, the marked plunge in the CSR has not drawn the kind of attention it deserves.
The decline in the CSR has been happening since 1961 although this time there was a sharp decline by 13 points. [In 2001, the all-India CSR was 927 girls for every 1,000 boys. It fell to 914 in 2011.] It could have been even sharper if it were not for the concern highlighted in Census 2001, following which the spotlight fell on the high-focus States, that is, those States considered as the worst offenders. Surprisingly, the attention given to these States made them improve their record slightly. Punjab, Haryana and Tamil Nadu showed improvement. Had it not been so, the graph would have plummeted further.
We have always prided ourselves that we present a snapshot, a camera picture, of the ground reality. We are not biased by programmes or policies. We never say “why”. What we would like to continue to say is that ‘this is the position'; policymakers, who are programme implementers, ‘please take note and do whatever corrections are required'. We have taken a very conscious decision on this. People ask me to comment on the reasons but I do not do that as we would become biased in the future. I would like to tweak the data to suit my hypothesis. We observe. Yes, the trend is very bad and it is a cause for concern. After 1991, a huge amount of concern emerged and it got picked up. Everyone did their bit.
Do you think village-level data, as and when they emerge, will help guide policymakers more effectively than State- or district-level aggregates? For example, it is quite possible that a few districts with positive CSRs can tilt the overall figure for the State.
As you drill down, it gives one more focussed intervention. It helps policymakers concentrate in those pockets where the issue is more rampant and pinpoint their responses. Resources can also be deployed more effectively. That is what the Census does, by going down to the village level. It gives a basis for informed decision-making.
In the recently released Sample Registration System (SRS) data, which show a modest decline in the infant mortality rate (IMR), the maternal mortality ratio (MMR), the under-five child mortality rate (U5MR) and fertility rates, the Millennium Development Goals (MDG) have been taken as the benchmark in the context of these mortality indicators.
The data vindicate what is happening on the ground. SRS is a sample survey, and one gets the same kind of results that the Census shows. It doesn't say that fertility has reached below replacement levels. We have also not said that population has stabilised. Fertility declines happen over a long period of time. To expect fertility declines every year is unrealistic. It is a national policy to view MDG goals; these goals are sacrosanct and we are striving towards achieving them. Three of the MDG goals happen to come from the SRS, so we have tried to highlight that there has been some progress towards them.
What are the challenges you anticipate with regard to the government's decision to enumerate castes, other than the Scheduled Castes and the Scheduled Tribes, between June and December 2011? There were objections to a caste Census on various grounds. It was suggested that a mere headcount was not enough to integrate caste data with socio-economic and demographic data, such as data for literacy, education, marital status and occupation.
It is not my area to question why it should be done. The government had a policy since Independence that caste enumeration would not be a part of the Census. Even in the 1931 Census, when it was done, and in the post-Independence phase, every Census Commissioner had expressed reservations about adding this question to the Census.
The government felt that there was a danger that the integrity of the Census numbers might be compromised as this was a highly emotive issue where people could be mobilised. Census tends to ask those questions which do not offer any incentives. We do not ask things like income where there is an incentive to give a misleading answer.
Despite the Census deploying a huge number of people, it is not possible to verify whatever everyone says. We take whatever is said as the truth. We do a post-enumeration survey, not a verifying exercise, going into the quality and content errors and then we declare a certain margin of error. But here a certain kind of authenticity is required and accurate numbers are needed first. After that, we could go in for the composition of the numbers by caste. This is not a count of only a particular category but of all castes. Every person can provide his/her caste name or can refuse to do so. Every person's caste is being enumerated and the names of all these castes are being taken.
The People of India project done by the Anthropological Survey of India has listed more than 70,000 names, which are highly complex. Many people write gotra, lineage and sub-caste as their caste. It is not possible for a part-time enumerator to quickly go around and collect all data accurately. Many people feel that a column could have been provided and data collected.
The government has decided that once caste data are collected, they will be placed before the nodal Ministries concerned – the Ministry of Tribal Affairs and Ministry of Social Justice and Empowerment. They will set up an expert committee and classify these returns. There are no lists that are common across the country. Many States do not have a Backward Castes list. A certain amount of expertise is needed to classify all this. The challenge is in conducting the field operations. We are using State government machinery for this.
What is the latest on the NPR? The Census office was required to collect data for this as well.
These are two independent exercises. The mechanism of the Census was used to collect data required for the NPR during the first phase of the Census, when we did the house listing round. At that a time a form was canvassed for the NPR. The data have been collected and all the 30-odd crore forms have been scanned and work has started in almost 13 or 14 States and more than a million people's biometric records have been collected. These records would be sent to the UID [unique identification] authority and Aadhaar numbers will be generated by them. It will come back to us. After that, the proposal is to put it on cards and issue the identity card to every person, which will be a smart card. Every resident of the country will get a smart card, which will be called the NPR resident's identity card.
This organisation [the office of the Registrar General] used to do only one Census once in 10 years; we are already doing three, the Census in two phases, the NPR and the caste census. We are contacting a billion people three times and that too on a run. We are putting all our might [into it].
‘We present a snapshot'
T.K. RAJALAKSHMI
Interview with Dr C. Chandramouli, Registrar General of India and Census Commissioner.
V. SUDERSHAN
C. Chandramouli, Census Commissioner, addressing the media after releasing the maternal mortality estimates in New Delhi on July 7.
IT was with eager anticipation that the results of the 15th provisional national Census (the 15th since 1872) were awaited. Census 2011 has thrown up some interesting and also worrying results: literacy rates have gone up; fertility levels have declined; the child sex ratio (CSR) has worsened since the last Census, with several districts reporting regressive trends; the growth rate of the population in the Empowered Action Group (EAG) States has registered a significant decline; and the population of children (0-6 years) has declined. While other sociocultural and demographic data, including data on housing and household amenities, are yet to come, the Census office launched its caste Census in June following a government decision taken last year. Census 2011 also laid the foundation for the National Population Register (NPR). The caste enumeration exercise is expected to be over by the end of the year though it may take another six months for the findings to get tabulated and collated before being released.
Dr C. Chandramouli, the Registrar General of India and Census Commissioner, expressed concern over the declining CSR and spoke about the challenges of caste enumeration, among other things, in an interview to Frontline. Excerpts:
The provisional Census has described the decline in the CSR as an all-time historical low. The Census results provide policymakers with a direction to undertake follow-up action. While the improvement in literacy rates and the decline in fertility levels were widely welcomed, the marked plunge in the CSR has not drawn the kind of attention it deserves.
The decline in the CSR has been happening since 1961 although this time there was a sharp decline by 13 points. [In 2001, the all-India CSR was 927 girls for every 1,000 boys. It fell to 914 in 2011.] It could have been even sharper if it were not for the concern highlighted in Census 2001, following which the spotlight fell on the high-focus States, that is, those States considered as the worst offenders. Surprisingly, the attention given to these States made them improve their record slightly. Punjab, Haryana and Tamil Nadu showed improvement. Had it not been so, the graph would have plummeted further.
We have always prided ourselves that we present a snapshot, a camera picture, of the ground reality. We are not biased by programmes or policies. We never say “why”. What we would like to continue to say is that ‘this is the position'; policymakers, who are programme implementers, ‘please take note and do whatever corrections are required'. We have taken a very conscious decision on this. People ask me to comment on the reasons but I do not do that as we would become biased in the future. I would like to tweak the data to suit my hypothesis. We observe. Yes, the trend is very bad and it is a cause for concern. After 1991, a huge amount of concern emerged and it got picked up. Everyone did their bit.
Do you think village-level data, as and when they emerge, will help guide policymakers more effectively than State- or district-level aggregates? For example, it is quite possible that a few districts with positive CSRs can tilt the overall figure for the State.
As you drill down, it gives one more focussed intervention. It helps policymakers concentrate in those pockets where the issue is more rampant and pinpoint their responses. Resources can also be deployed more effectively. That is what the Census does, by going down to the village level. It gives a basis for informed decision-making.
In the recently released Sample Registration System (SRS) data, which show a modest decline in the infant mortality rate (IMR), the maternal mortality ratio (MMR), the under-five child mortality rate (U5MR) and fertility rates, the Millennium Development Goals (MDG) have been taken as the benchmark in the context of these mortality indicators.
The data vindicate what is happening on the ground. SRS is a sample survey, and one gets the same kind of results that the Census shows. It doesn't say that fertility has reached below replacement levels. We have also not said that population has stabilised. Fertility declines happen over a long period of time. To expect fertility declines every year is unrealistic. It is a national policy to view MDG goals; these goals are sacrosanct and we are striving towards achieving them. Three of the MDG goals happen to come from the SRS, so we have tried to highlight that there has been some progress towards them.
What are the challenges you anticipate with regard to the government's decision to enumerate castes, other than the Scheduled Castes and the Scheduled Tribes, between June and December 2011? There were objections to a caste Census on various grounds. It was suggested that a mere headcount was not enough to integrate caste data with socio-economic and demographic data, such as data for literacy, education, marital status and occupation.
It is not my area to question why it should be done. The government had a policy since Independence that caste enumeration would not be a part of the Census. Even in the 1931 Census, when it was done, and in the post-Independence phase, every Census Commissioner had expressed reservations about adding this question to the Census.
The government felt that there was a danger that the integrity of the Census numbers might be compromised as this was a highly emotive issue where people could be mobilised. Census tends to ask those questions which do not offer any incentives. We do not ask things like income where there is an incentive to give a misleading answer.
Despite the Census deploying a huge number of people, it is not possible to verify whatever everyone says. We take whatever is said as the truth. We do a post-enumeration survey, not a verifying exercise, going into the quality and content errors and then we declare a certain margin of error. But here a certain kind of authenticity is required and accurate numbers are needed first. After that, we could go in for the composition of the numbers by caste. This is not a count of only a particular category but of all castes. Every person can provide his/her caste name or can refuse to do so. Every person's caste is being enumerated and the names of all these castes are being taken.
The People of India project done by the Anthropological Survey of India has listed more than 70,000 names, which are highly complex. Many people write gotra, lineage and sub-caste as their caste. It is not possible for a part-time enumerator to quickly go around and collect all data accurately. Many people feel that a column could have been provided and data collected.
The government has decided that once caste data are collected, they will be placed before the nodal Ministries concerned – the Ministry of Tribal Affairs and Ministry of Social Justice and Empowerment. They will set up an expert committee and classify these returns. There are no lists that are common across the country. Many States do not have a Backward Castes list. A certain amount of expertise is needed to classify all this. The challenge is in conducting the field operations. We are using State government machinery for this.
What is the latest on the NPR? The Census office was required to collect data for this as well.
These are two independent exercises. The mechanism of the Census was used to collect data required for the NPR during the first phase of the Census, when we did the house listing round. At that a time a form was canvassed for the NPR. The data have been collected and all the 30-odd crore forms have been scanned and work has started in almost 13 or 14 States and more than a million people's biometric records have been collected. These records would be sent to the UID [unique identification] authority and Aadhaar numbers will be generated by them. It will come back to us. After that, the proposal is to put it on cards and issue the identity card to every person, which will be a smart card. Every resident of the country will get a smart card, which will be called the NPR resident's identity card.
This organisation [the office of the Registrar General] used to do only one Census once in 10 years; we are already doing three, the Census in two phases, the NPR and the caste census. We are contacting a billion people three times and that too on a run. We are putting all our might [into it].
