A hypothesis on the monitoring system in India’s Total Sanitation Campaign

Printer-friendly versionPrinter-friendly version

One of the problems of the Total Sanitation Campaign (TSC) in India has been its flawed monitoring system. The sanitation sector internationally was shocked when recently the sanitation coverage data from the Census were published. The worst fears were surpassed. Between 2001 and 2011, the TSC reported a sanitation increase of 46 points; from 22% to 68%. However, coverage was only 31% in 2011 according to the Census (GoI 2012a). This raises questions about the TSC monitoring system –and about the TSC policy itself, too.

One of these questions is: what does TSC monitoring system monitor? While writing a paper on the TSC (forthcoming), Brian Bell and I tried to examine this question and set up a hypothesis: It monitors reported execution, irrespective of what actually happens on the ground. That is, it measures what those in the administration in charge of the TSC report to have been executed. The data are poured into the TSC online monitoring system at the local level, without much checking of what has actually happened on the ground.

We had local level evidence that backed our hypothesis, but wanted to explore if our hypothesis applied to the system as a whole.

Therefore, we elaborated a graph with the reported toilet coverage according to the TSC online monitoring system –green line– (GoI 2011) and the toilet coverage with the toilet coverage according to the Census and other surveys –red points and red tendency line– (GoI 2012a; IIPS 2007, 2010). We then started digging in the TSC online monitoring data and found out that the reported coverage was very much related to the amount of funds spent (GoI 2012b). Specifically, if we take 2000 as the joint starting point, we see that the curve representing the cumulative funds released by the centre for the TSC –black– runs almost parallel to the reported coverage.

 

This reinforces our idea that TSC monitoring system is about reported execution, irrespective of actual outcomes.

Sadly, the recently revamped Indian sanitation policy does not address this issue, which has implications not only for the monitoring system, but also for a sanitation policy that claims to be demand led but appears to be money driven and supply led.

References:

Government of India (GoI) (2012a). Census of India 2011. Houses, Household Amenities and Assets. Latrine Facility. Government of India.

GoI (2012b). Country level financial year-wise release/expenditure amounts. Retrieved 09/18, 2012, available from the Accountability Initiative here

GoI (2011). Enhanced Quality of Life through Sustained Sanitation. India Country Paper. SACOSAN - IV edn. Colombo, Sri Lanka: Department of Drinking Water and Sanitation. Ministry of Rural Development. Government of India. Available at: http://www.indiaenvironmentportal.org.in/files/India%20Country%20Paper%202011.pdf

Hueso, A. and Bell, B. (forthcoming). An untold saga of policy failure: the Total Sanitation Campaign in India

International Institute for Population Sciences (IIPS). (2010). District Level Household and Facility Survey (DLHS-3), 2007-08: India. Mumbai: IIPS.

IIPS. (2007). National Family Health Survey (NFHS-3), 2005-2006: India. Mumbai: IIPS.

Date: 29 November 2012
Contributors: 
Region: 
Country: 
India

Comments

Submitted by Andy Robinson (not verified) on

A good point well made. But the situation varies significantly across States. I made a similar analysis of TSC progress data against household survey sanitation data for WSP in 2010, and found that the TSC data in Himachal Pradesh was reasonably accurate (household survey latrine usage data showing approximately 10%-20% lower latrine usage than reported in the TSC data), but that the TSC data in Madhya Pradesh was highly inaccurate (household surveys reporting 80% lower latrine use rates than TSC). Unsurprisingly, the validity of the TSC data depends largely on the policy and political commitment of the State and local governments! More detail here:

http://www.wsp.org/sites/wsp.org/files/publications/WSP-India-Endline-EE.pdf    

Submitted by Andrés Hueso (not verified) on

Thank you very much for your comment, Andy, and for the link, took. I read your work before and found it very interesting! I have indeed made my PhD research in the two states you mention and you are right that the situation varies a lot between states. However, I believe that my point can be generalised to the TSC as a whole without much bias. In further analysis I have done comparing TSC and Census data (see here), the most populated states show higher levels of over-reporting (red colour). Himachal Pradesh does much better, as you pointed, although over-reporting was still quite higher than expected. Regarding the causes of this, political economy issues were the most relevant aspects affecting over-reporting (which are very closely related to political commitment, yes).

Submitted by Greg Pierce (not verified) on

Hi Andres,

Thanks for your blog posts on the Total Sanitation Campaign. They are very insightful.

I have used the 2007-2008 DLHS data to build a simple spatial regression model of toilet use at the district level. I'd be interested to hear your thoughts on the quality of the DHLS data as well as see an extended version of your paper 'An untold saga of policy failure: the Total Sanitation Campaign in India'

Please contact me at your convenience. Best,

 

Greg Pierce

gspierce@ucla.edu

Doctoral Candidate

Department of Urban Planning

University of California, Los Angeles