India's upkeep of datasets is extremely callous!
Conducting research in India has always been challenging as the quality of data is neither dependable nor reliable. A glaring example of this is the discrepancy in the data of Indian GDP for the first quarter of 2010-11 released on August 31, 2010. The Central Statistical Office (CSO), an official unit that keeps track of India’s various economic indicators, showed 8.8 per cent growth in GDP at factor cost, which was higher to market price by 3.65 per cent. The error happened because of the usage of wrong deflators — it was corrected the next day by placing the GDP growth rate (at market price) to 10.1 per cent.
Within a few weeks of this faus pax, another error occurred in data dissemination. The Index of Industrial Production (IIP) indicated an astronomical growth — from 7.1 per cent in June to 13.8 percent in July! Further, IIP growth in August was shown at 5.6 per cent y-o-y while most other analysts felt it would be 10 per cent. There has also been inconsistency between the estimates of national income in National Accounts Statistics (NAS) and estimates of per capita consumption by National Sample Survey Organisation (NSSO). The data on rural poverty by NAS shows that it is about 10 to 15 per cent – in sharp contrast to NSSO consumption data that shows rural poverty hovering around 41.8 per cent! These are recent case; but there are several other crucial socio-economic indicators wherein accurate data is hard to come by. A case in point is inflation! Be it the wholesale price index (WPI) or various consumer price indices, neither of these gives the right picture of inflation. Inflation based on WPI fell sharply last year, whereas the price of fruits, sugar, and milk soared at the retail level. That’s a clear paradox! The WPI went negative, while food prices sky rocketed!
Access to accurate data is just one side of the problem. Most of these inaccuracies also emanate from the fact that India’s official database suffers from multiple defining parameters too. It’s no hidden truth that India itself has multiple poverty and per capita definition, resulting in multiple data when it comes to poverty estimations. This not only appears confusing to any external analyst but also provides a provision for policy makers to both use and abuse these indicators at will. Moreover, there are multiple agencies working on same set of databases with practically no coordination and them. Unlike other developed nation, we still don’t have a centralised system of maintaining data. Furthermore, our official statistic database is rarely updated on a regular basis and is generally full of errors.
On one hand the data is inaccurate and non-dependable, while on the other there is no credible source of information. There are no reliable statistical figures on service sector, which accounts for more than half of the country’s GDP and no updated records on unemployment. Also, when it comes to extrapolating data of unorganised sector’s contribution to the GDP, there are no recent data available. The parameters taken in such instances are old and not periodically updated.
Such a poor record of important indicators is definitely a matter of utter shame. But it is more embarrassing to witness the manner in which policy makers take advantage of these inaccuracies! It is time that data collection and dissemination is outsourced to independent credible agencies, bereft of any influences, which would also be open to all kinds of scrutiny to prove the veracity of information.
Conducting research in India has always been challenging as the quality of data is neither dependable nor reliable. A glaring example of this is the discrepancy in the data of Indian GDP for the first quarter of 2010-11 released on August 31, 2010. The Central Statistical Office (CSO), an official unit that keeps track of India’s various economic indicators, showed 8.8 per cent growth in GDP at factor cost, which was higher to market price by 3.65 per cent. The error happened because of the usage of wrong deflators — it was corrected the next day by placing the GDP growth rate (at market price) to 10.1 per cent.
Within a few weeks of this faus pax, another error occurred in data dissemination. The Index of Industrial Production (IIP) indicated an astronomical growth — from 7.1 per cent in June to 13.8 percent in July! Further, IIP growth in August was shown at 5.6 per cent y-o-y while most other analysts felt it would be 10 per cent. There has also been inconsistency between the estimates of national income in National Accounts Statistics (NAS) and estimates of per capita consumption by National Sample Survey Organisation (NSSO). The data on rural poverty by NAS shows that it is about 10 to 15 per cent – in sharp contrast to NSSO consumption data that shows rural poverty hovering around 41.8 per cent! These are recent case; but there are several other crucial socio-economic indicators wherein accurate data is hard to come by. A case in point is inflation! Be it the wholesale price index (WPI) or various consumer price indices, neither of these gives the right picture of inflation. Inflation based on WPI fell sharply last year, whereas the price of fruits, sugar, and milk soared at the retail level. That’s a clear paradox! The WPI went negative, while food prices sky rocketed!
Access to accurate data is just one side of the problem. Most of these inaccuracies also emanate from the fact that India’s official database suffers from multiple defining parameters too. It’s no hidden truth that India itself has multiple poverty and per capita definition, resulting in multiple data when it comes to poverty estimations. This not only appears confusing to any external analyst but also provides a provision for policy makers to both use and abuse these indicators at will. Moreover, there are multiple agencies working on same set of databases with practically no coordination and them. Unlike other developed nation, we still don’t have a centralised system of maintaining data. Furthermore, our official statistic database is rarely updated on a regular basis and is generally full of errors.
On one hand the data is inaccurate and non-dependable, while on the other there is no credible source of information. There are no reliable statistical figures on service sector, which accounts for more than half of the country’s GDP and no updated records on unemployment. Also, when it comes to extrapolating data of unorganised sector’s contribution to the GDP, there are no recent data available. The parameters taken in such instances are old and not periodically updated.
Such a poor record of important indicators is definitely a matter of utter shame. But it is more embarrassing to witness the manner in which policy makers take advantage of these inaccuracies! It is time that data collection and dissemination is outsourced to independent credible agencies, bereft of any influences, which would also be open to all kinds of scrutiny to prove the veracity of information.
Excellent article! But credibility of independent commercial agencies is equally doubtful! Recent IDC, GFK handset sales data are proof of that. Yes, credibility can improve if the agencies are really independent and free from unwarranted influence.
ReplyDeleteSir ...thanks for sharing this with us!!!!
ReplyDelete