Open government (data)

Comment is free, but facts are sacred

Thanks to the exceedingly good central government run website to file applications under the Right to Information Act (see end of this post for details), I got the opportunity to look at some useful data on implementation of large nationwide schemes. I have been trying to obtain data on such schemes across subjects, disciplines and departments with the objective of understanding what ails the management and utilisation of data in government services in India. Take for example the poor utilisation of routine data on births and deaths, which is of immense importance to policymaking. Our births and deaths registration system needs serious overhauling, if it has to do more than give a birth and death certificate. My earlier experience with trying to understand how our centuries-old heritage animal collections are maintained was also not very different. Over the next few months, more of my data-based test-RTIs are anticipated and I am hoping will tell me a different story than my previous (lack of) data experiences across other departments, including environment and forests, tribal affairs, health and labour.

About RSBY from ruby.gov.in
About RSBY from rsby.gov.in

It was a chance discussion with a colleague of mine that triggered my interest in the Rashtriya Swasthya Bima Yojana, RSBY for short. The RSBY has been hailed as a landmark “cashless and paperless” health financing scheme that sought to ensure that people below poverty line do not end up facing the catastrophic effects of health expenditure, let alone the illness itself. Several researchers have shown the individual, familial and social burden of health expenditure on the poor, healthcare often bought in the open market at the cost of education, employment or other essential activities that feed healthcare expenditure among the poor.

Anyway, to cut a long story short RSBY is one of those large well-intentioned nationwide schemes in the league of the National Rural Health Mission and the MNREGA. Such schemes are implemented in a complex setting that our country is  (or any other one for that matter). With the involvement of private for-profit insurance companies, third party agencies, state and district governments and several other actors across the several states of the country, the need for continuous monitoring and evaluation, and course correction from time to time is indispensable.  However, a routine scan of government websites shows very scant data and information in public domain. While websites themselves are very poorly designed on one hand, the data available about the various schemes remains at the level of mentioning monies spent rather than assessing effectiveness. See for example Victoria Fan and Ajay Mahal’s article in the National Medical Journal of India on “Learning and getting better: Rigorous evaluation of health policy in India“.

There have of course been evaluations by independent researchers and by the government itself (see official commissioned studies here). See for example this study that assesses the financial protection provided by RSBY in Gujarat, or this proposed study with an interesting experimental trial design. Such evaluation studies are resource-intensive and their scope is regional. Whereas there is a mountain of routine data that ought to be in the public domain, analysed, discussed, critiqued and hopefully fed back into implementation and policy. Alas, that is not very common. For example, in response to a question in Parliament in 2012, the (then) labour Minister gave a few numbers (see here) which is part of the little nationwide RSBY implementation data that is in public domain. One would anticipate a nationwide landmark scheme such as this would have a more robust monitoring system that put data out in public domain so that the discourse about the scheme happens in public domain.

Section 4 of the RTI requires all government departments to make available their routine data in public domain through computerisation.
Section 4 of the RTI requires all government departments to make available their routine data in public domain through computerisation.

Indeed, if not for the right to information act, a lot of such data would have been completely inaccessible to the citizens. In fact, the section 4 of the RTI directs government departments to statutorily disclose a lot of information that is nowadays being subject to unnecessary RTI. One would imagine that the scheme website would allow a journalist, researcher or a concerned citizen to query their implementation database and say download the enrolment of people, utilisation and monies disbursed in a given geographical region. Alas, that is not the case.

In response to my RTI on enrolment and utilisation of the RSBY scheme, I got very preliminary data on enrolment of people below poverty line across Indian states. In response to my query on numbers of people who “utilised” the scheme, I was given the data on “number of smart cards issued”. Apparently, the respondent of my RTI assumed that issue of smart cards is “utilisation” of the scheme, much reflecting prevailing attitudes of assuming money allocated to a scheme as being a proxy for implementation/utilisation. I am hoping to appeal and get the utilisation data in public domain as well, but the larger question is that of open government. The first step in understanding the implementation and effectiveness of such large schemes should be to make the data available in public domain. As you can see below, the data I ask for are extremely basic programme data that ought to be easily accessible.

RTI application to Ministry of Labour

That said, the Government of India’s online RTI portal by the Department of Personnel & Training, is one of the best government websites I have ever seen. There are a few difficulties in logging in and the site is a bit slow, but the transition from filing of RTIs to obtaining responses is so seamless and well thought through. The site itself is very simple; after initial registration (not mandatory), one can apply online and make the payment of Rs. 10 using their online payment facility. Be aware though, that the website is strictly for central government departments and no RTI pertaining to any state is allowed. Once you file the RTI, the updates are integrated with your phone so that your receive regular SMS updates and in most cases, the contact details (including phone number and emails sometimes) of the public information officer handling your application. The portal organises transfer of application from one department to another in case that is needed. Altogether a wonderful experience. Many other such RTI pulls of potentially public data are on the cards. 🙂

The RTI portal of Government of India is a very well implemented website with seamless access to information from all central government departments

Meanwhile, I did get some enrolment data. Some preliminary analysis shows (as yet) poor enrolment of people below poverty line in many states (note that many states that show ZERO enrolment may not actually be implementing the scheme at all, for example Tamil Nadu & Andhra Pradesh). I have used the targeted BPL families as the denominator to calculate enrolment rates. Most of their enrolment targets (denominators) came from the RSBY website itself. In essence, what these trends show is the extent to which the poor are being enrolled (utilisation and coverage is still another story) into the scheme. While increasing in most states, the enrolment coverage is largely poor. If RSBY is to be a part of any of the new government’s universal health coverage plan, it is not yet there. And any such plans have to seriously look into the performance of the scheme in various settings before adopting the RSBY-route.

Data based on enrolment figures provided by Ministry of Labour in response to RTI. Denominators to calculate enrolment percentages based on targeted BPL/poor as per RSBY website and other sources. All years indicate financial years (2008 means 2008-09). Dip in 2013 could be due to data not yet being updated.
Data based on enrolment figures provided by Ministry of Labour in response to RTI. Denominators to calculate enrolment percentages based on targeted BPL/poor as per RSBY website and other sources. All years indicate financial years (2008 means 2008-09). Dip in 2013 could be due to data not yet being updated.

Please see corrected version of the plot with data for Maharashtra included here.

The R code used to generate the graph is here


Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.