Bibek Debroy
Pity that some things have to be taken seriously and can’t be dismissed as nonsense. I am talking about the Report of the Committee for Evolving a Composite Development Index of States, otherwise known as the Raghuram Rajan Committee. This is not a purely academic exercise and will have policy implications. Inter-State rankings are a dime a dozen. But this one is different, or ought to be. It’s certainly different in the slipshod and whimsical way it has gone about it, despite the apparently impeccable academic credentials of the authors. Let me separate this Committee’s devolution recommendation requirements from the ranking and the index. In this blog, I will only focus on the ranking and the index. In any inter-State ranking, you have to do the following. (a) Identify the variables – often paucity of data prevents you from including variables you might otherwise have wished to include. (b) Figure out a method of normalization. (c) Assign weights to the identified variables. (d) Decide on a method of aggregation. These are serious issues and do realize any tinkering with any of these changes the value of the index and the consequent ranking. Therefore, you need to give justifications for your answers to (a) to (d) and play around with some kind of simulation exercise, to check how robust your index values and rankings are.
The Committee accepted ten clusters of variables: (1) income; (2) education: (3) health; (4) household amenities index; (5) poverty ratio; (6) female literacy rate; (7) % of SC/ST population; (8) urbanization rate; (9) financial inclusion; and (10) connectivity index. Perhaps these are the variables to include, perhaps not. Perhaps there are other variables. Why are we doing this? To ensure that citizens throughout the country have access to the same quality of public goods and services. If there are States, where a substantial segment of the population lives in villages that have population sizes less than 500 and if there is hilly and difficult terrain too, the costs of delivery will be higher. So should that not be a criterion? Is SC/ST categorization per se an indicator of backwardness, or is there that apparent causal relation because of inadequate access to public services? For instance, there is some research for SC populations, where if you control for those other factors, being per se SC is not an inherent disadvantage. ST populations are somewhat different, because they tend to be concentrated in parts of the country where there are problems with both physical and social infrastructure. Should I take the female literacy rate or should I take the female work participation rate? Monthly per capita consumption expenditure is taken as a surrogate indicator for income. Should I take that, or should I take a specific component of that, such as private consumption expenditure on non-food items, or on education and health? Should one consider own task revenue as a share of GSDP (gross State domestic product)? What about use of LPG? On connectivity, I would have thought pucca roads is a very good indicator. In financial inclusion, why only banks? Why not post offices? What about number of policemen? I am not suggesting that you take one or the other of these variables, or that you reject the cluster of variables accepted by the Committee. I am making the point that there should be some discussion for acceptance and rejection. Do you find it? The answer is no.
Consider now the data sources. Take education. This comes from NSS. The point about NSS is that large-size NSS sample surveys occur typically at intervals of five years. Should one base identification on data that will become available after 5 years? For school education, I would have got a decent set of data, every year, from DISE. In addition, data on several variables come from Census. That means data will only be available after 10 years. The one that really puzzled me was infant mortality rate, the only indicator for health. It puzzled me because the last health survey was in 2005-06. I looked up the data source and it says SRS Bulletin, October 2012. That’s fine. SRS Bulletin does have IMR data for States. But let’s also be clear, these are estimates. It may be fine to use estimates. However, academic rigor would have required you to state that these are estimates, not actuals based on surveys. I don’t find that mentioned anywhere in the report. That’s on variables. What about normalization? It’s by no means obvious that normalization by population is necessarily superior to normalization by geographical area. On weights, PCA (principal components analysis) is superior because the weighting system is generated from the exercise. We have been told equal weights have been used, because those are easier to understand. I don’t think facility of understanding should have been a criterion. Assuming it is a criterion, the present exercise isn’t particularly easy to understand. It smacks of being arbitrary and subjective.
Had this come to me as a paper submitted for publication and sent to me for "refereeing”, I would have sent it back, asking for substantial revisions. It is a least developed report. And the tragedy is that, unlike a journal paper, there will be serious discussions around it.
Source: http://blogs.economictimes.indiatimes.com/policypuzzles/entry/raghuram-rajan-committee-report-on-states-least-developed
Pity that some things have to be taken seriously and can’t be dismissed as nonsense. I am talking about the Report of the Committee for Evolving a Composite Development Index of States, otherwise known as the Raghuram Rajan Committee. This is not a purely academic exercise and will have policy implications. Inter-State rankings are a dime a dozen. But this one is different, or ought to be. It’s certainly different in the slipshod and whimsical way it has gone about it, despite the apparently impeccable academic credentials of the authors. Let me separate this Committee’s devolution recommendation requirements from the ranking and the index. In this blog, I will only focus on the ranking and the index. In any inter-State ranking, you have to do the following. (a) Identify the variables – often paucity of data prevents you from including variables you might otherwise have wished to include. (b) Figure out a method of normalization. (c) Assign weights to the identified variables. (d) Decide on a method of aggregation. These are serious issues and do realize any tinkering with any of these changes the value of the index and the consequent ranking. Therefore, you need to give justifications for your answers to (a) to (d) and play around with some kind of simulation exercise, to check how robust your index values and rankings are.
The Committee accepted ten clusters of variables: (1) income; (2) education: (3) health; (4) household amenities index; (5) poverty ratio; (6) female literacy rate; (7) % of SC/ST population; (8) urbanization rate; (9) financial inclusion; and (10) connectivity index. Perhaps these are the variables to include, perhaps not. Perhaps there are other variables. Why are we doing this? To ensure that citizens throughout the country have access to the same quality of public goods and services. If there are States, where a substantial segment of the population lives in villages that have population sizes less than 500 and if there is hilly and difficult terrain too, the costs of delivery will be higher. So should that not be a criterion? Is SC/ST categorization per se an indicator of backwardness, or is there that apparent causal relation because of inadequate access to public services? For instance, there is some research for SC populations, where if you control for those other factors, being per se SC is not an inherent disadvantage. ST populations are somewhat different, because they tend to be concentrated in parts of the country where there are problems with both physical and social infrastructure. Should I take the female literacy rate or should I take the female work participation rate? Monthly per capita consumption expenditure is taken as a surrogate indicator for income. Should I take that, or should I take a specific component of that, such as private consumption expenditure on non-food items, or on education and health? Should one consider own task revenue as a share of GSDP (gross State domestic product)? What about use of LPG? On connectivity, I would have thought pucca roads is a very good indicator. In financial inclusion, why only banks? Why not post offices? What about number of policemen? I am not suggesting that you take one or the other of these variables, or that you reject the cluster of variables accepted by the Committee. I am making the point that there should be some discussion for acceptance and rejection. Do you find it? The answer is no.
Consider now the data sources. Take education. This comes from NSS. The point about NSS is that large-size NSS sample surveys occur typically at intervals of five years. Should one base identification on data that will become available after 5 years? For school education, I would have got a decent set of data, every year, from DISE. In addition, data on several variables come from Census. That means data will only be available after 10 years. The one that really puzzled me was infant mortality rate, the only indicator for health. It puzzled me because the last health survey was in 2005-06. I looked up the data source and it says SRS Bulletin, October 2012. That’s fine. SRS Bulletin does have IMR data for States. But let’s also be clear, these are estimates. It may be fine to use estimates. However, academic rigor would have required you to state that these are estimates, not actuals based on surveys. I don’t find that mentioned anywhere in the report. That’s on variables. What about normalization? It’s by no means obvious that normalization by population is necessarily superior to normalization by geographical area. On weights, PCA (principal components analysis) is superior because the weighting system is generated from the exercise. We have been told equal weights have been used, because those are easier to understand. I don’t think facility of understanding should have been a criterion. Assuming it is a criterion, the present exercise isn’t particularly easy to understand. It smacks of being arbitrary and subjective.
Had this come to me as a paper submitted for publication and sent to me for "refereeing”, I would have sent it back, asking for substantial revisions. It is a least developed report. And the tragedy is that, unlike a journal paper, there will be serious discussions around it.
Source: http://blogs.economictimes.indiatimes.com/policypuzzles/entry/raghuram-rajan-committee-report-on-states-least-developed
No comments:
Post a Comment