WATER AND SANITATION PROGRAM: REPORT 76714 Global Scaling Up Sanitation Project Linking Service Delivery Processes and Outcomes in Rural Sanitation: Findings from 56 Districts in India The Water and Sanitation Program is a multi-donor partnership administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. This assessment was undertaken by a team from the Water and Sanitation Program (WSP), which included C. Ajith Kumar, Upneet Singh, Suseel Samuel, Mariappa Kullappa, Snehal Shah, Rajiv Raman, Dipak Roy, Manu Prakash, Kakumanu Arokiam, Aravinda Satyavada, Rama Rohini Sarvepalli, Megha Jain, and Amarendra Kumar. Statistical analysis and development of graphs for this report were undertaken by Snehal Shah and Upneet Singh. Acknowledgement is due to the participants of the various workshops where this study was presented, whose feedback and questions led to an enrichment of the contents of this report. The final report was peer reviewed by William D. Kingdom, Lead Water and Sanitation Specialist, and Srinivas Podipireddy, Senior Water and Sanitation Specialist (both SASDU) and Susanna Smets, Senior Water and Sanitation Specialist (WSP, East Asia). Thanks are also due to Christopher Juan Costain, Regional Team Leader (WSP), for guidance and feedback on successive drafts, Vandana Mehra, Communications Officer (WSP), for support in the design, production, and dissemination of this study, and Lira Suri, Team Assistant (WSP), for coordinating logistics and backup support during the study period. The Ministry of Drinking Water and Sanitation (MDWS), Government of India (GoI), is also acknowledged for sharing the data set and findings of the GoI study on sustainability of sanitation behavior in local governments that had been awarded the Nirmal Gram Puraskar or Clean Village Prize. This data set forms a major source of information for this report. We would like to thank Mr. J.S. Mathur, (former) Joint Secretary to GoI, and Mr. Vijay Mittal, (former) Director, MDWS, for their feedback, guidance, and encouragement throughout the process of undertaking this assessment. While all efforts have been made to ensure that the data presented are correct, any inadvertent errors remain the responsibility of the author team. Team Leader: C. Ajith Kumar Task Manager: Upneet Singh Written by: Upneet Singh, Ajith Kumar, Snehal Shah, Suseel Samuel, and Mariappa Kullappa Peer Reviewers: Susanna Smets, William Kingdom, and Srinivas Podipireddy Editor: Ananda Swaroop Design: Colorcom Advertising The Water and Sanitation Program is a multi-donor partnership administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. Rights and Permissions: Water and Sanitation Program (WSP) reports are published to communicate the results of WSP’s work to the development community. Some sources cited may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed herein are entirely those of the authors and should not be attributed to the World Bank or its affiliated organizations, or to members of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank Group concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to wsp@worldbank.org. WSP encourages the dissemination of its work and will normally grant permission promptly. For more information, please visit www.wsp.org. ©2013 Water and Sanitation Program Global Scaling Up Sanitation Project Linking Service Delivery Processes and Outcomes in Rural Sanitation: Findings from 56 Districts in India Contents Executive Summary .................................................................................3 Chapter 1: Introduction ............................................................................7 1.1 Background................................................................................7 1.2 Objectives ..................................................................................9 1.3 Conceptual Framework..............................................................9 1.4 Methodology ............................................................................13 1.5 Limitations................................................................................21 Chapter 2: Correlating Service Delivery Processes with Outcomes....23 2.1 How have Sample Districts Scored on Quality of Processes? ..............................................................23 2.2 Does Quality of Service Delivery Processes Matter?...............24 Chapter 3: What Explains the Difference in District Scores on Service Delivery Processes? ...............................................27 3.1 Catalyzing Program Implementation........................................28 3.2 Implementing Program Goals ..................................................30 3.3 Sustaining Outcomes...............................................................32 Chapter 4: Identifying Priorities for Sector Reforms ............................36 4.1 Sustaining Outcomes...............................................................36 4.2 Catalyzing Outcomes ...............................................................37 4.3 Implementing Outcomes..........................................................38 References ......................................................................................39 Annex 1: Selection of Household Sample by State and District .............40 Annex 2: Interview Guide ........................................................................42 Abbreviations and Acronyms ................................................................47 Glossary ......................................................................................48 List of Tables 1. Sample States and Number of NGPs won between 2005-08 .16 2. Selection of Sample Districts and NGP Winners ....................16 3. Types of Stakeholders and Number of Interviews at District Level ............................................................................17 4. Instruments for Data Collection on Processes ........................18 5. Process Rating Scale ...............................................................19 6. Results of Individual Process Correlation with Usage Outcomes .....................................................................26 7. Prioritizing Processes for Reform ...........................................36 List of Figures 1. Framework for analyzing processes linked with performance .............................................................................10 2. Institutional structure for GoI’s Total Sanitation Campaign .....13 3. Study methodology..................................................................15 4. How have sample districts scored on quality of processes in terms of components? .............................................................23 www.wsp.org 1 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Executive Summary 5. How have sample districts scored on quality of processes? ..24 6. Correlating quality of processes with sustaining Nirmal status ............................................................................24 7. Correlating sample district scores on process components with district performance in terms of toilet usage...............................................................................25 8. 56 districts’ average scores on quality of processes...............27 9. Comparing district scores: Strategy ........................................28 10. Comparing district scores: Institutions ....................................28 11. Comparing district scores: Financing ......................................29 12. Comparing district scores: Demand ........................................30 13. Comparing district scores: Supply and Technology Promotion ................................................................................31 14. Comparing district scores: Scaling up .....................................31 15. Comparing district scores: Subsidy Delivery ...........................33 16. Comparing district scores: Monitoring ....................................33 17. Comparing district scores: Rewards........................................34 List of Boxes 1. Evolution of Rural Sanitation Programs in India .......................8 2. Nirmal Gram Puraskar: Efforts to Strengthen the Monitoring System .....................................................................9 3. Key Definitions ........................................................................10 4. GoI Assessment of NGP Impact and Sustainability ................14 5. Study Sample...........................................................................18 6. Examples of Other Studies that Assess Service Delivery Processes ................................................................................22 7. Institutional Model to Scale up Total Sanitation: Example of a District that has a High Score on Quality of Processes ...........29 8. Promoting Sanitation as Toilet Construction, not Behavior Change ...............................................................31 9. Whose Toilet? Lack of Household Involvement in Selecting Toilet Technology .....................................................32 10. Phased Approach to TSC Implementation in Raigad District, Maharashtra .............................................................................32 11. Subsidy as Unit Cost for Selecting Toilet Model......................34 12. Cross-verification of TSC Monitoring Data ..............................35 13. Non-monetary Incentives: Example of Dakshin Kannada District in Karnataka .................................................35 Numbers 1 lakh 100,000 1 million 1,000,000 1 crore 10,000,000 1 billion 1,000,000,000 Currency and Exchange Rate US$ 1 = INR 51 (April 2012 exchange rate) 2 Global Scaling Up Sanitation Project Executive Summary The year 2012 marked the close of the Government of the findings from 56 detailed district reports on the qual- India’s (GoI’s) Total Sanitation Campaign (TSC). The ity of service delivery processes adopted. It uses a ‘process program was started in 1999 to achieve universal rural sani- rating scale’ to identify areas of improvement and gaps in tation coverage. TSC was given a fillip with the introduc- the delivery of the rural sanitation program. District scores tion of an incentive program known as the Nirmal Gram on the process rating scale are correlated with district Puraskar (NGP, Clean Village Prize) in 2003. NGP offered performance in terms of toilet usage to understand whether a cash prize to motivate local governments at village, block, these factors are correlated and to what extent. Based on and district levels to achieve total sanitation. Although rural this analysis, the assessment identifies priorities for reforms sanitation coverage has increased during TSC, progress still and proposes practical solutions to facilitate the adoption of falls short of the program goal of universal coverage. Further, processes linked with better performance. national aggregates also conceal differences between, and within, states and districts in terms of performance on TSC Assessing the Quality of Processes goals. In this context, the question of which service deliv- In the first round of this assessment undertaken in 2010, ery processes are linked with better performance assumes six service delivery processes were identified as being critical significance. for achieving sustainable rural sanitation outcomes at scale. In this second round, the number of processes assessed was To address this question, the Water and Sanitation Program increased to nine and grouped into three thematic compo- (WSP), in partnership with the Ministry of Drinking nents corresponding to the three stages in which the districts Water and Sanitation (MDWS), undertook an assessment implement the program: Catalyzing, Implementing, and of processes that drive outcomes in sector performance. A Sustaining (see Figure 1). This is based on the logic that first round of the assessment was undertaken in 2010 which an enabling environment is first required to catalyze imple- covered 22 districts in 21 states (GoI and WSP 2010), mentation and focus on the achievement of behavior change followed by this second round in 2011 which covered 56 outcomes. Once this is in place, program implementation districts in 12 states. needs to focus on aligning operations to facilitate achieve- ment of outcomes. Finally, the outcomes achieved must be TSC is currently being restructured as the Nirmal Bharat sustained to achieve downstream impacts such as improve- Abhiyan (NBA, Clean India Campaign). This provides a ments in health and quality of life. timely opportunity to revisit what is working well and how this can be replicated. It is equally important to identify Sample districts’ percentage scores on the quality of the the bottlenecks that are holding back progress and deter- processes were found to have a significant and positive corre- mine how these can be addressed. This assessment distils lation with the extent of toilet usage in rural households. FIGURE 1: FRAMEWORK FOR ANALYZING PROCESSES AND PERFORMANCE COMPONENT I: COMPONENT II: COMPONENT III: CATALYZING IMPLEMENTING SUSTAINING RURAL Strategy Demand Subsidy Delivery SANITATION Institutions Supply Monitoring OUTCOMES Finance Scale Rewards and Incentives www.wsp.org 3 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Executive Summary The basis for calculating rural toilet usage was the percent- To illustrate the differences in the quality of processes adopt- age population among NGP-winning villages, aggregated ed by districts, one high performer and one low performer at the district level, reporting regular usage of household were selected from the sample on the basis of their perfor- toilets and safe disposal of child feces. The source of this mance in terms of toilet usage. The approach adopted by information is a national assessment of NGP sustainabil- these two districts on each of the nine processes tracked in ity undertaken by MDWS in 2010 (GoI 2010). In other the rating scale is summarized in Table 1. words, higher the quality of processes adopted at the district level, stronger the likelihood that the population in Identifying Priorities for Reform NGP-winning villages will report sustained behavior change Among the three components analyzed, Sustaining had the linked to household toilet usage and safe disposal of child strongest correlation with usage/behavior change outcomes feces. District scores on the three individual components but the second lowest rating score. This indicates that the of the process rating scale—Catalyzing, Implementing, and program was relatively more successful in putting in place Sustaining—were also found to have significant positive a framework for service delivery in terms of strategy, insti- correlations with district-level usage outcomes. tutional structure, and financial allocation for rural sanita- tion. However, translating these enabling conditions into Quality of Processes: service delivery outcomes and sustaining these outcomes How have Sample Districts Scored? was a challenge for program implementers. Among the three thematic components, sample districts scores on the quality of processes tended to be the strongest Sustaining outcomes on Catalyzing, followed by Implementing and Sustaining. Monitoring had the strongest correlation with outcomes in This means that the districts performed well in terms of this component but the second lowest process rating score putting in place policies, institutions, and budgets related among all the nine processes tracked. The present monitor- to program implementation. However, it has been challeng- ing system of TSC focuses on inputs and outputs achieved ing to translate these enabling conditions into scaling up in the short-term (toilet construction) rather than sustain- demand and supply to reach and sustain outcomes. ing behavior change (toilet usage). The NGP monitoring system looks at community-wide Open Defecation Free Looking at the nine individual processes in the rating scale, (ODF) status but since it is designed as a one-time prize, we find that districts scored the strongest on supply, financ- the program’s focus shifts from the winners to those Gram ing, and institutional setups. Scores were average on all other Panchayats (GPs) that are still eligible to win. Surveys of processes except for rewards and recognition, on which sanitation behavior in rural areas make information avail- district scores were the lowest among all processes tracked. able in the public domain every three to four years by when it is often too late to make a mid-course correction. One Program, Diverse Results: A mechanism is needed between these extremes to ensure What do High Scorers do Differently? that timely and reliable information is available to check the In order to understand the factors that underlie differential pulse of the program at periodic intervals. scores on quality of service delivery processes in districts that are implementing a common national rural sanitation Some recommendations for reform, looking at the oppor- program, the sample of 56 districts was distributed into tunity provided by the Nirmal Bharat Abhiyan, include: three categories—high scorers, average scorers, and low • The proposal to release the NGP prize money in two scorers—in descending order of their scores on the qual- installments, with the second installment released ity of processes. The largest gaps between high and low subject to Nirmal status being sustained for at least scorers, in each component of the rating scale, were found one year needs to be operationalized. Re-survey of in strategy under Catalyzing, in demand creation under previous years’ winners can be undertaken by survey Implementing, and in monitoring under the Sustaining agencies along with current year’s applicants; component. • Some states (Maharashtra, Himachal Pradesh, and 4 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Executive Summary TABLE 1: COMPARISON OF APPROACHES ADOPTED BY HIGH PERFORMING AND LOW PERFORMING DISTRICTS ON NINE TRACKED PROCESSES Component Process High Performing District Low Performing District (toilet usage is >90%) (toilet usage is <35%) Strategy Focus on collective behavior change Focus on toilet construction rather to use toilets rather than practice open than toilet usage defecation CATALYZING Institutions Dedicated staff appointed and efforts Sanitation competes with other devel- made to strengthen capacity at district opment programs for priority; many and sub-district levels to implement the vacancies despite sanctioned posts rural sanitation campaign Finance Funding used for both short- and long- Funding routed to meet short-term term goals; households motivated to toilet construction targets invest own funds Demand creation All Information, Education and IEC material developed but dissemi- Communication (IEC) channels used nation is a weak link effectively; focus on behavior change IMPLEMENTING Supply Menu of technology options promoted Incentive under TSC interpreted as which reflect consumer choices and unit cost, used to determine single affordability technology model Scaling up Phased implementation in geographic Lack of strategic direction; targets set and thematic terms, e.g., move from open at centralized levels are often discon- defecation free (ODF) to total environmen- nected from ground situations tal sanitation Subsidy Delivery Program guidelines implemented in letter Subsidy is used to construct toilets and spirit; subsidy given as incentive without commensurate efforts at demand creation SUSTAINING Monitoring Focus on outcomes and quality of data Focus on inputs (budget spent) and reported; NGP winners are periodically outputs (toilets constructed); little verified effort to track sustainability in NGP- winning local governments Rewards Institute competition-based awards, Main incentive meeting input and monetary and non-monetary prizes and output targets honors to prioritize sanitation Karnataka, to name a few) have also introduced makes data available online and is updated on a annual competitions between GPs to be recognized monthly basis. This can be further strengthened by as the cleanest. This means that GPs need to not only utilizing mobile-to-web technology which has the sustain community-wide ODF status but also exceed potential to make information on outcomes avail- their Nirmal status; and able at shorter intervals compared to conventional • The TSC Management Information System (MIS) pen and paper surveys. www.wsp.org 5 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Executive Summary With respect to the delivery of subsidy, there are propos- of the resources to focus on achievement of outcomes rather als to increase the amount of subsidy available and also to than short-term targets. make Above the Poverty Line (APL) families eligible for this under the Nirmal Bharat Abhiyan. This assessment’s find- Implementing Outcomes ings show that there is still some way to go in terms of ensur- The priority for reform in implementation is processes relat- ing that the subsidy is utilized as an incentive rather than a ed to demand creation, which had the lowest process rating unit cost. Instead of focusing on the amount of subsidy per score in the implementing component. There is a misper- se, the focus can be on ensuring that this is released after ception that demand can be created after toilet construc- behavior change and not against toilet construction. tion. On the contrary, demand creation begins with entry into a community and the process adopted to achieve sani- Catalyzing outcomes tation goals at entry. Also, once toilet construction targets Although strategy was found to have the stronger correla- are met and reflected in routine monitoring, incentives to tion with outcomes, this was also the process with the lowest ensure the quality of construction and motivate behavior process rating score within this component. A key weakness change are reduced. While sample districts scored well over- of the current program is that although districts (and states) all on supply of sanitation products and services, assessment have the flexibility to develop a strategy that responds to findings show that, in many cases, there was a tendency to the local context and capacity, and even though the guide- promote a single technology model which fit the subsidy lines advocate behavior change and community-wide ODF available for construction of toilets. This is not conducive status, most tend to follow the national campaign guide- for demand generation or promoting behavior change. lines in letter rather than in spirit. In terms of institutions, Finally, in terms of scale, expecting motivators to work on capacity needs to be strengthened. There is a need to create a a voluntary basis is not sustainable and training must be dedicated sanitation cell at the district level which includes accompanied by provisions to ensure that they are provided specialists with expertise in monitoring, communications, with logistical and other support to work across commu- and so on, to manage the campaign. In addition, there is a nities. A key area of improvement for districts is ensuring need to strengthen capacity at sub-district levels as well. In prioritization of the program in terms of selection of villag- financing, the challenge for districts is to make effective use es as well as behaviors. 6 Global Scaling Up Sanitation Project I. KEY POINTS Introduction • The Census of 2011 finds that one in three rural house- holds had access to a toilet, up from one in five in 2001. However, national performance aggregates conceal disparities between and within states and districts that are implementing a common national rural sanitation program; and • Service delivery processes that underlie differential achievements of states and districts are tracked only to a limited extent. This assessment is an attempt to unpack the ‘process black box’ to gain insights into which processes are linked with better (or worse) performance. This chapter provides the background of the study and the of the cash prize varied from ` 50,000 (approximately rural sanitation sector in India. It defines key concepts— US$1,000) to ` 500,000 (approximately US$ 10,000) service delivery processes and outcomes—and sets out depending on the population size. The prizes were given the rationale for the selection of a district as the unit of once a year to qualifying local governments. enquiry. The Methodology section explains key sources of information used, basis for sample selection, methods for Rural sanitation coverage increased during TSC and the measurement of service delivery processes and outcomes, Census of 2011 found that one in three rural house- and limitations faced. holds had access to a toilet, up from one in five in 2001. However, the scale of the sanitation challenge in India is 1.1 Background such that, despite this improvement, the country remains The rural sanitation sector in India is at a turning point. home to the largest number of people practicing open Despite the implementation of two large national rural defecation in the world (JMP 2010). It is estimated that sanitation programs to date by the Government of India one in every 10 deaths in India is linked to poor sanitation (GoI) (Box 1), the goal of an open defecation free (ODF) and hygiene. Diarrhea, a preventable disease, accounts for India—a Nirmal Bharat1—has remained elusive. The year every 20th death. Prevalence of child under nutrition in 2012 marked the close of a second national rural sanitation India (47 percent according to National Family Health program, the Total Sanitation Campaign (TSC), launched Survey-III, 2005-06) is among the highest in the world in 1999 to achieve universal rural sanitation coverage by and nearly double that in Sub-Saharan Africa.3 If the 2012, a goal that has now been deferred to 2022.2 TSC economic losses linked to poor sanitation are monetized, was provided a fillip with the introduction of an incentive the results are staggering. In Purchasing Power Parity program known as the Nirmal Gram Puraskar (NGP) or (PPP) terms, adverse economic impacts of inadequate Clean Village Prize in 2003. The NGP program awarded- sanitation in India are US$161 billion, or US$144 per cash prizes to motivate local governments at village, block, person. This is the equivalent of 6.4 percent of the Gross and district levels to achieve total sanitation. The amount Domestic Product (GDP) or ` 2.4 trillion in 2006 (WSP 2010). 1 Literally, totally sanitized India. Nirmal here refers to total sanitation as defined by the eligibility criteria of the Nirmal Gram Puraskar (Clean Village Prize) of the Govern- In response to this sanitation challenge, GoI has made ment of India. To qualify as Nirmal, a local government should be 100 percent free from open defecation at the household and institutional levels and have systems in place significant investments in rural sanitation. Two issues for safe disposal of solid and liquid waste. 2 3 The goal of achieving universal rural sanitation coverage in India coincides with the In 2006, 43.5 percent of Indian children under five were underweight for their age end date of the Government of India’s national rural sanitation program. The first compared to 16.4 percent in Uganda and 13.9 percent in Ghana. This finding is not national rural sanitation program, the Central Rural Sanitation Program, started in a result of different measurement standards or ethnic factors, and studies have shown 1986 and set the goal at 1999. However, the program was not successful in achieving that, given similar opportunities, children across ethnic groups can grow to the same universal rural sanitation coverage and was restructured as TSC in 1999, which set the levels. Three reasons are highlighted: low birth weight; lower status of women in India goal at 2012. While this study was being compiled, TSC was in the process of being in decision-making which limits their control over resources related to nutrition; and restructured as the Nirmal Bharat Abhiyan (Clean India Campaign) which defers the poor sanitation and hygiene which is a leading cause of infections related to under date for achieving universal rural sanitation coverage to 2022. nutrition in under-twos (Gragnolati et al. 2005). www.wsp.org 7 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction BOX 1. EVOLUTION OF RURAL SANITATION PROGRAMS IN INDIA Central Rural Sanitation Program (1986-98): The Central Rural Sanitation Program (CRSP) was GoI’s first national rural sanitation initiative. It was launched in 1986 and focused on providing 100 percent hardware subsidy to generate demand for pour-flush toilets. Commensurate efforts to motivate toilet usage or promote technology options were absent (GoI 2008). CRSP failed to motivate or sustain sanitation coverage, and the Census of 2001 found that 78 percent of rural households in India practiced open defecation (Registrar General of India 2002). In light of its relatively poor performance, GoI restructured CRSP at the national level and launched TSC in 1999. Total Sanitation Campaign (1999-2012): TSC advocated a participatory and decentralized approach. It moved away from the infrastructure-focused approach of the earlier program and concentrated on promoting behavior change. Some key features of TSC include: • A community-led approach with focus on collective achievement of total sanitation; • Focus on Information, Education, and Communication (IEC) to mobilize and motivate communities towards safe sanita- tion; • Minimum capital incentives only for Below the Poverty Line (BPL) households, post construction and usage. Later the program introduced post-achievement incentives—NGPs—for local governments to achieve total sanitation; • A flexible menu of technology options; and • Development of a supply chain to meet the demand stimulated at the community level. Sources: GoI (2008); Registrar General of India (2001). can be highlighted here. Firstly, national performance which defers the goal of universal rural sanitation to the aggregates conceal significant disparities in performance year 2022. There is growing recognition that a ‘business as between states and districts that are implementing rural usual’ approach will not be sufficient to achieve this goal. sanitation initiatives under common national program The monitoring system of the incentive program, NGP, is guidelines. Secondly, GoI’s routine monitoring system is also being constantly improved through annual reviews of geared towards periodic tracking of inputs (budget spent) the selection procedure and guidelines (Box 2) to preserve and outputs (toilets constructed). Since the goal of the the prestige of this award and motivate local governments program is an ODF India, focusing on construction of to take the lead in changing their sanitation status. latrines and disbursement of financial allocations skews the incentives for implementers in favor of prioritizing To address the question of which processes drive district building of toilets rather than their use. Outcomes such as performance in rural sanitation, WSP, in partnership with ODF communities are tracked to a limited extent through the Ministry of Drinking Water and Sanitation (MDWS), the NGP verification process. Therefore, while it is possi- undertook an assessment of the processes that drive ble to identify best (and worst) performers in terms of outcomes in sector performance. A first round of the assess- inputs, outputs, or outcomes, the question that remains ment was undertaken in 2010 which covered 22 districts unanswered is: which service delivery processes are linked to in 21 states5 followed by this second round in 2011 which better performance? covered 56 districts in 12 states. The target audience for this assessment is policy makers and implementers at national This question gains importance in view of the ongo- and state levels. ing restructuring of TSC into the Nirmal Bharat Abhiyan (Clean India Campaign),4 a third national program, 5 4 In the Indian context, the use of the word ‘campaign’ denotes the national rural GoI and WSP (2010): A Decade of the Total Sanitation Campaign: Rapid Assess- sanitation program. ment of Processes and Outcomes, Vols. 1 and 2. 8 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction BOX 2. NIRMAL GRAM PURASKAR: EFFORTS TO STRENGTHEN THE MONITORING SYSTEM Effective monitoring is the backbone of a successful incentive program. It ensures that only those local governments that actually deserve the award are recognized, maintaining the integrity and prestige of the award as a whole. Based on the experience of operationalizing NGP, the incentive program’s guidelines have been revised by GoI in successive years to strengthen the verification process. Some of the changes introduced in the verification process include: • Collection of identification data for each respondent household; this can be voter identification, a ration card number, or other identification. This is done to ensure that the existence of that household within the boundaries of the applicant local government can be proven, if challenged; • All NGP applicants are verified by independent survey agencies. To further strengthen the quality of verification, 30 percent of the applicants qualifying for NGP are cross-checked by a team deputed from another state; • Release of prize money in two installments such that the first installment is released immediately upon qualifying for the prize. The second installment is released after six months subject to qualifying in a re-verification of 100 percent ODF status; and • Concurrent monitoring of the verification process by international agencies such as UNICEF, Water Aid and the Water and Sanitation Program (WSP) of the World Bank. Sources: GoI (2010) and WSP (2010). 1.2 Objectives 1.3.1 What is meant by service delivery processes and The objectives of this assessment are to: components? • Provide GoI with an overview of the range of service In the first round of this assessment, six processes were delivery processes adopted by different districts identified as being critical to achieving sustainable rural across states and their relative performance in terms sanitation outcomes at scale: 1) strategy; 2) institutional of outcomes; framework; 3) approach to demand creation; 4) technol- • Identify which service delivery processes are linked ogy promotion and supply chain; 5) monitoring; and 6) to better (or worse) performance; and financing. The rating scale used in the first round gave • Identify where the key bottlenecks are to achiev- equal weight to the different process factors. However, a ing or sustaining outcomes—are they in catalyzing key point of the feedback from policy makers and practi- program execution, in implementing the program tioners on the findings of the first round of this assessment and its goals or in sustaining outcomes? was that all factors did not influence performance equally. Hence, those that had a larger bearing on the direction or 1.3 Conceptual Framework rate of achievement of outcomes should receive a higher This assessment is based on the hypothesis that the quality of score. Also, since GoI had invested in two large national service delivery processes by which a district implements a rural rural sanitation programs, each of which spanned a period sanitation program determines the outcomes achieved and their of 12 years, factors related to sustainability of outcomes sustainability. At the outset, the two principal components were more important at this stage of the program compared of this hypothesis are defined: to those linked to starting the program and catalyzing its • Service delivery processes and components; and implementation. • Outcomes (Box 3). Accordingly, the second round of this assessment used a The unit of inquiry is a district and this section explains revised rating scale to measure the quality of processes. The why a district was selected as the basis for comparison of number of processes assessed was increased from six to nine service delivery processes and outcomes. with the addition of the following three processes: mode of www.wsp.org 9 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction FIGURE 1: FRAMEWORK FOR ANALYZING PROCESSES LINKED WITH PERFORMANCE COMPONENT I: COMPONENT III: COMPONENT II: SUSTAINING CATALYZING IMPLEMENTING RURAL • Subsidy Delivery • Strategy • Demand SANITATION • Monitoring • Institutions • Supply OUTCOMES • Rewards and • Finance • Scale Recognition subsidy delivery;6 scaling up; and rewards and recogni- BOX 3. KEY DEFINITIONS tion. These three additional processes were added based on Service Delivery Processes: These refer to the steps the feedback from policy makers and practitioners on the taken by district governments to achieve sanitation first round of this assessment which pointed out that these outcomes under GoI’s flagship rural sanitation program, service delivery processes should be assessed individually TSC (restructured as Nirmal Bharat Abhiyan). Projects rather than as a part of larger processes as they are critical outside the purview of TSC are not included in this for the achievement of outcomes. These nine processes were assessment. This study simplifies the measurement grouped into three thematic components corresponding to of these steps by categorizing them into nine proc- the three stages by which districts implement the program: esses. These processes are further grouped into three Catalyzing, Implementing, and Sustaining (Figure 1). This thematic components—Catalyzing, Implementing, and was based on the logic that first an enabling environment Sustaining—depending on the stage of the service is required to catalyze the implementation of a sanitation delivery cycle to which they correspond (Figure 2 and program and to focus on the achievement of outcomes. linked explanation). Once this is in place, program implementation needs to focus on service delivery to achieve outcomes. And finally, Outcomes: In this study, outcomes are defined as the outcomes achieved must be sustained to achieve down- usage of toilets by rural households. Information stream impacts such as improvement in health and quality on outcomes was sourced through a GoI assess- of life. ment on NGP sustainability (details in section 1.4 on Methodology). This assessment included a survey A brief description of the processes analyzed in this assess- of 9,960 rural households where respondents were ment (by component) is given below. For each process, the asked, for each member aged four years or more, if he/ explanation as outlined in the national program guidelines she regularly went for open defecation or used a toilet. is provided. This is the normative explanation or what the For children up to three years, feces disposal methods program design should look like if it follows the national were classified as safe (disposed in toilet/buried) or guidelines in letter and in spirit. The focus of this assess- unsafe (open dumping). The number of persons using ment is on the extent to which actual district processes align a toilet and the number of children under three whose with or diverge from the principles and approach contained feces were disposed safely together constituted the in the national guidelines. ODF population which was converted into a percent- 6 Here, a distinction is made between ‘mode of subsidy delivery’ and ‘financing’. Fi- age of the total population surveyed to arrive at the nancing refers to the program budget and allocation for various components, for ex- district-level outcome number. ample, IEC, hardware incentive, and so on, and processes related to release and expen- diture. Mode of subsidy delivery refers to the process by which the hardware incentive is delivered to a household that is entitled to it. 10 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction Component 1: Catalyzing timing, and sequence. The TSC Guidelines advo- • Strategy: The TSC Guidelines provide a broad frame- cate a demand-driven approach to rural sanitation work within which states and districts have the flex- backed by post-achievement incentives. Districts ibility to devise their own strategy for implementing have the flexibility to implement this principle based the program at scale, depending on the socio- on their context and capacity. economic and institutional context, terrain, and • Technology Promotion and Supply Chain: The TSC capacity. Overall, the guidelines advocate a partici- Guidelines advocate informed technology choice patory and community-led approach to rural sanita- and setting up of alternate supply channels such as tion which focuses on behavior change to end open Rural Sanitary Marts (RSMs) to facilitate supply of defecation. To implement this approach in letter and products and services, if required. At the implemen- in spirit, an understanding of the program’s strategy tation level, technology promotion includes not only at different levels (district headquarters to the village) separate toilet components (for example, sanitary is required along with political and administrative pans, pipes, traps, and so on) but also existing latrine will to prioritize program implementation based technology options (for example, septic tanks, venti- on participatory principles in the face of compet- lated double-pit toilets, ecological sanitation, and so ing priorities. The focus of this assessment is the on). It also includes provision of masonry services extent to which the strategy to implement TSC at for installation and sanitary services for operation, the district level follows the principles and approach maintenance, and final disposal. Hence, districts advocated by the national guidelines. have the flexibility to promote a range of technology • Institutions and Capacity: Institutions define the options and supply chains for products and servic- framework for service delivery. To set the stage for es under TSC, and the pathways adopted by them delivering program goals, a nodal agency needs to to support such technology promotion and supply be identified as the institutional home for TSC with chain development are examined through this assess- clear roles, responsibilities, and resources to fulfill its ment. mandate. Within the institutional framework, space • Scaling up: By design, TSC is a national campaign needs to be made to allow local governments at the and implemented at scale in all rural districts of village level to play a key role as advocated in the India. This process has, therefore, been included TSC Guidelines. Capacity refers to the availability not with respect to geographical scale but in terms of skilled human resources for program implementa- of the quality of the approach to planning adopted tion. According to the guidelines, provisions can be by a district for scaling up. Scaling up as a participa- made for trained technical and administrative staff tory approach requires engaging trained motivators to be available at different levels to coordinate the and providing them with incentives and/or logistical program. The extent to which districts comply with support to reach all the villages in a district, spread- this guidance is assessed through this study. ing implementation across phases such that lessons • Finance: Financing refers to budgetary allocations learnt can be fed back into implementation, and to finance program activities. This includes costs setting clear priorities instead of tackling all program of activities under different program components components at once. (for example, software activities, school sanitation and hygiene education, administration, and so on) Component 3: Sustaining as well as the process by which funds are allocated, • Subsidy Delivery: Under TSC financing guidance, released, and spent. the government’s contribution to individual house- hold toilets for Below the Poverty Line (BPL) house- Component 2: Implementing holds may be extended if this is deemed necessary • Creating Demand and Community Mobilization: A for the complete involvement of the community to program approach consists of specific activities, their reach ODF status. The contribution is intended as www.wsp.org 11 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction an incentive rather than subsidy and, therefore, the years. This was based on the assumption that rural house- guidelines specify that ‘the construction of a toilet holds are too poor to afford the cost of constructing a toilet should be undertaken by a BPL household itself, and hence subsidies were provided for toilet construction in and on completion and use of the toilet by the BPL rural areas. Often, toilets provided through this approach household, cash incentive can be given in recogni- were either not used or used for other purposes such as stor- tion of its achievement’ (TSC Guidelines 2007: p. age or bathing.7 This approach underwent a paradigm shift 8). This assessment examines how the TSC incen- with the restructuring of CRSP into TSC (1999-2012) tive is treated by a district, that is,as an incentive which introduced a focus on motivating behavior change or reward for behavior change as intended by the and ODF communities. guidelines or as a subsidy for toilet construction. It also examines the process by which the incentive or 1.3.3 Why look at district-level processes? subsidy is targeted and delivered to rural households. To understand why a district is selected as the basis of this • Monitoring: Large-scale sanitation programs such assessment, it is essential to first understand the institution- as TSC require an efficient monitoring system and al structure for rural sanitation. At the national level, the the ability to ensure that the results of monitor- Department of Drinking Water and Sanitation (DDWS)8 of ing are used to improve program implementation. the Ministry of Rural Development (MoRD) is responsible Since TSC advocates a community-wide approach, for rural sanitation. In most states, the rural sanitation port- the extent to which districts track sanitation cover- folio is given either to the Rural Development Department age in Above the Poverty Line (APL) households (RDD) or the Public Health Engineering Department that are not covered by the TSC incentive/subsidy (PHED). Program management arrangements also provide is examined. This study also looks at the extent to suitable institutional arrangements at the district and sub- which district-level monitoring systems are able to district levels (Figure 2). capture, process, and use monitoring data to improve program implementation and accurately report on TSC operates through district projects of three to five years’ program progress. duration, each jointly financed by GoI, the state govern- • Rewards and Recognition: Incentives can be mone- ment, and beneficiary households. The funding split varies tary or non-monetary based on recognition or other according to the program component but averages about 65 intangible benefits. Incentives and disincentives can percent from GoI, 23 percent from the state government, motivate program implementers to comply with key and 14 percent from beneficiaries. Fund flow for centrally principles based on a ‘carrot and stick’ approach. This (GoI) sponsored programs such as TSC takes place directly assessment looks at the extent to which districts have to the district, with the matching state share being released introduced incentives or disincentives to sustain the when utilization begins and a request is put in by the motivation of different actors (political leadership, district. A district has to submit a project proposal (in line program staff at the headquarters and at the field with the guidelines issued by GoI) and a project implemen- level, communities, and so on). tation plan to the state government to avail of GoI funding. Each state is required to establish appropriate institution- 1.3.2 What is meant by outcomes? al arrangements to facilitate project implementation by In the context of this study, outcomes are defined as usage districts and monitor implementation. Specialist consul- of toilets by rural households. Hence, districts with higher tants from the fields of communication, human resource toilet usage by rural households are assessed as having better 7 Although intended as an incentive, hardware subsidies tend to thwart the goal of toilet outcomes compared with districts that have lower toilet usage by shifting the focus to toilet construction; the latter does not automatically lead usage. In the Indian rural sanitation context, toilet usage is to the former. The Union Minister for Drinking Water and Sanitation, Jairam Ramesh, summarized the gap between toilet construction and usage as: ‘I personally have seen distinct from toilet coverage. Due to the legacy of programs a large number of toilets in different parts of the country which are not being used for such as CRSP (1986–98), the sector was dominated by a toilets. They are being used as storage god owns’ (cited in The Hindu). 8 While this assessment was underway, DDWS was hived off from MoRD and con- construction-driven approach to rural sanitation for many verted into MDWS. 12 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction FIGURE 2: INSTITUTIONAL STRUCTURE OF TOTAL SANITATION CAMPAIGN 1.4 Methodology In the Indian rural sanitation sector, information on toilet usage is collect- CENTER Government of India Funding, technical support, ed to a limited extent through the (Ministry of Drinking Water and M&E, training and Sanitation) inter-sectoral coordination annual NGP prize verification survey but the results are not available in the public domain. Survey data are avail- able every three to four years9 but Funding, technical support, these are usually outdated. Hence, STATE State Government development of state action at the outset, a reliable and relatively (Nodal Department) plan, inter-sectoral coordination, training, M&E recent source of outcome data was required with which process ratings could be correlated. Unit of implementation DISTRICT under TSC. Facilitates Zila Panchayat overall implementation, In 2010, GoI initiated a national survey (and other government and non-government institutions) development of action plan, to assess the extent of sustainability of inter-sectoral coordination, training, M&E Nirmal status in NGP-winning Gram Panchayats (GPs, village-level local governments, Box 4), including a Panchayat Samiti Institution building (e.g., sample survey of rural households to BLOCK GPs, watsan committee), (Extension workers of the facilitate supply chains, ascertain the extent of toilet coverage government and non-government hygiene education, organizations) monitoring and usage in local governments that had won the NGP between 2005 and 2008.10 The household survey data Institution building, set, including findings related to toilet VILLAGE Gram Panchayat mobilization, facilitating the usage by rural households, was made (Motivators) construction of hardware, Community hygiene education, available by the Ministry to WSP in monitoring, O&M January 2011 to access information on outcomes. Accordingly, the same M&E: Monitoring and Evaluation; Watsan: Water and Sanitation; O&M: Operations and Maintenance. 56 district sample across 12 states which was selected for GoI’s study development (HRD), monitoring, and school sanitation of NGP sustainability was selected for this assessment of and hygiene education can be appointed as consultants at processes undertaken by WSP in partnership with GoI.11 the state level, and where required at the district level. Data on service delivery processes were collected between March and August 2011. States have the flexibility to devise their own specific An overview of the methodology and sources of data with approaches within the broad framework provided by time frames is shown in Figure 3. The remainder of this the program. Several states, notably Himachal Pradesh, section provides a detailed explanation of the methodology Haryana, and Maharashtra, have modified the TSC in the following steps: Guidelines to dovetail better with their existing sanita- tion initiatives. In addition, the TSC Guidelines provide 9 a flexible framework for the district projects, allowing For example, a District Level Health and Facility Survey is conducted at four-year intervals; the most recent round was concluded in 2008. different methodologies to be adopted depending on differ- 10 GoI (2011): Assessment of Impact and Sustainability of Nirmal Gram Puraskar. ent contexts, demands, and capacities within the district New Delhi: Ministry of Drinking Water and Sanitation. 11 Note that since TSC was implemented by all districts in India, this precluded the (Robinson and Raman 2007). possibility of undertaking a counter-factual analysis. www.wsp.org 13 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction BOX 4. GOI’S ASSESSMENT OF NGP IMPACT AND SUSTAINABILITY DDWS launched a comprehensive study to assess the impact and sustainability of the NGP incentive program in 2010. The main purpose of the study was to assess the impact of NGP on sanitation coverage and usage and its related impact on health, education, gender empowerment, and social inclusion in rural areas on different user groups, particularly rural poor. This study also focused on the durability and sustainability of the provision and usage of sanitary facilities over time. The focus areas identified for the study were: 1. Current Status of NGP criteria 1.1 Status of toilet coverage, durability and functionality. 1.2 Status of usage and Nirmal status (ODF and resolution, garbage disposal, and drainage systems). 2. Impact and Hygiene 2.1 Impact on health, education, economics, gender, and social inclusion. 2.2 Relation of impact with status of hygiene practices. 3. Sustainability of NGP Status 3.1 Critical factors for achieving NGP status. 3.2 Reasons for not building/using toilets among different target groups. 3.3 Factors critical for better sustainability of NGP status. 3.4 NGP award money utilization. 4. Measures and Modifications for Sustained NGP Status 4.1 Measures needed to strengthen the impact and sustainability. 4.2 Recommendations on whether to continue NGP and suggestions for improvement. Source: GoI(2011). 1.1.1 Sampling strategy: for outcome data collection household but whether it was used or not was not known 1.1.2 Sampling strategy: data collection on service as toilet usage is not tracked by GoI’s routine monitoring delivery processes system. The number of states selected and the number of 1.1.3 Instruments for data collection NGPs won by each state in the four years between 2005 1.1.4 Training of interviewers and collection of data (this is the first year when the annual NGP was given) to 1.1.5 Analysis: linking service delivery processes with 2008 are shown in Table 2. NGP winners from 2009 and outcomes thereafter were not selected as the focus of the GoI study was on sustainability in those GPs that had won the prize at 1.4.1 Sampling strategy for outcome data collection least two years prior to the study being initiated rather than The sampling strategy adopted by GoI’s assessment of NGP winners in more recent years.12 sustainability is reproduced below as this formed the basis of the sample that was used in this assessment of service As can be seen from Table 1, there were relatively few delivery processes. NGP awardees in 2005, 2006, and 2007, compared to 2008 which saw an upward spike in the number of NGP Selection of states. A total of 12 states were selected as the winners. To provide a balanced distribution, the number sample for the study, distributed equally across the three of NGP winners across the four years was divided into two categories of high, average, or low performers. In this case, time groups wherein the winners from 2005, 2006, and categorization of performance was based on the percentage 2007 were clubbed into one time group (hereafter referred of rural households reported to have a toilet as per the data 12 Although the sample NGP winners are selected from different years in which the available through GoI’s routine monitoring system. This prizes were given from 2005 onwards, the analysis is based on a single survey under- meant that a toilet facility had been constructed in that taken in 2010 and published in 2011 (GoI-CMS 2011). 14 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction FIGURE 3: STUDY METHODOLOGY Methodology Joint GoI + WSP GoI assessment asssessment of TSC of sustainability processes in the of NGP in 56 districts same 56 districts Year undertaken Year undertaken Sample - 9,960 2010, NGP winners Sample - 112 GPs 2011, 2 NGP winners households, 664 selected from two in 56 districts in selected in each GPs in 56 districts time groups: 12 states district, one from across 12 states pre-2008 and 2008 each time group Qualitative: Key person Household Focus group/ interviews with program staff at Quantitative: survey (9,960 key person state, district, block and village Process rating households)* interviews* level. District reports with scale analysis of processes % population using toilet (including District level safe disposal of child feces) Outcomes correlated with processes scores on in NGP-winning GPs, aggregated at processes district level converted into % *Only outcome (toilet usage) from the 9,960 household survey was used to correlate with process ratings. Data from focus group/key person interviews were not used in the assessment of processes. to as pre-2008) and winners from 2008 formed the second While implementing the sampling method, it was found time group. that in the case of Rajasthan, the pre-2008 time group had very few NGP winners. Hence, all pre-2008 NGP winning Selection of districts and NGP winners. Districts where GPs had to be included in the study, and instead of the allo- NGP winners were found in both pre-2008 and 2008 time cated three districts, five districts had to be selected to fulfill groups were identified as ‘common districts’ in each sample the prescribed coverage of 18 NGP-winning GPs from each state. Approximately 30 percent of the common districts, time group. Therefore, the total sample of 56 districts was that is, 54 districts out of the 176 common districts in the arrived at with two additional districts being selected from sample states were selected for the study. For sampling, Rajasthan. these 54 districts were proportionally allocated to each state after allowing for a minimum of two districts in a state. From these 56 sample districts, the Terms of Reference The allocated number of districts was selected for each state developed by GoI specified that six to seven NGP winners using the Probability Proportional to Size (PPS) technique, had to be selected from each time group in each district. taking the number of GPs as the size variable. This implies Hence, a total of 664 NGP-winning GPs was selected, that districts with a higher number of GPs had a higher comprising an equal member (332) from each time group. probability to be selected in the sample. The selection of NGP-winning GPs was done using the PPS www.wsp.org 15 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction TABLE 1: SAMPLE STATES AND THE NUMBER OF NGPS WON 2005-08 Performance State NGPs NGPs NGPs Pre-2008 2008 Total Category Won in 2005 Won in 2006 Won in 2007 NGP Winners NGP (2005+ Winners 2006+2007) High Kerala 1 6 220 227 600 827 West Bengal 10 126 470 606 328 934 Tripura 1 36 46 83 30 113 Haryana 0 0 60 60 798 858 Average Maharashtra 13 380 1,974 2,367 4,301 6,668 Uttar Pradesh 0 40 488 528 729 1,257 Himachal Pradesh 0 0 22 22 245 267 Andhra Pradesh 0 10 143 153 662 815 Low Rajasthan 0 0 23 23 141 164 Bihar 0 4 39 43 156 199 Chhattisgarh 0 12 90 102 300 402 Karnataka 0 0 121 121 479 600 Total   25 614 3,696 4,335 8,769 13,104 method and taking the population of NGP-winning GPs as of 24 NGP winners in the pre-2008 time group and seven the size variable. However, in some districts in Rajasthan, NGP winners in the 2008 time group was found. This Uttar Pradesh, Bihar, Tripura, Himachal Pradesh, and shortfall was supplemented through NGP winners from Haryana, fewer than six or seven GPs had won NGPs in other selected districts in the same state through fresh PPS at least one of the specified time groups. A total shortfall sampling. The total number of districts and NGP winners TABLE 2: SELECTION OF SAMPLE DISTRICTS AND NGP WINNERS Category State No. of No. of No. of NGP Winning No. of NGP-GPs No. of NGP-GPs Common Districts GPs in Common Selected for Pre-2008 Selected for 2008 Districts* Allocated Districts* Time Group Time Group Pre-2008 2008 Per Total Per Total District District High Kerala 13 4 228 548 6 24 6 24 West Bengal 13 4 606 327 6 24 6 24 Tripura 3 2 79 30 7 14 7 14 Haryana 8 3 60 295 6 18 6 18 Average Maharashtra 33 9 2,367 4,301 6 54 6 54 Uttar Pradesh 54 15 528 688 6 90 6 90 Himachal Pradesh 5 2 22 105 7 14 7 14 Andhra Pradesh 17 5 152 662 6 30 6 30 Low Rajasthan 8 5 23 141 6 18 6 18 Bihar 7 2 44 156 7 14 7 14 Chhattisgarh 5 2 102 300 7 14 7 14 Karnataka 10 3 71 479 6 18 6 18 Total 176 56 4,282 8,769 76 332 76 332 * Common districts here refer to those districts where NGP winners were found in both pre-2008 and 2008 time groups. 16 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction from each time group selected in each sample state is shown six or seven villages selected for the GoI assessment of NGP in Table 2. sustainability in each district. Selection of households. House listing was carried out in Selection of respondents for interviews. In each district, the selected 664 NGP-winning GPs and 15 households semi-structured, individual interviews were conducted with from each NGP winner were selected using the Circular key actors. The criteria used to select actors for interviews at Systematic Sampling procedure. Hence, a total of 9,960 the district level included: households was selected for household interviews (for the • Should have been engaged in implementing the number of sample households selected in each state and rural sanitation program for at least six months; and district see Annex 1). • Should represent a key decision-maker or imple- menter at the district level. 1.4.2 Sampling strategy: data collection on service delivery processes The types of stakeholders interviewed and the number of Selection of states and districts. The sample of 12 states and interviews done are given in Table 3. 56 districts selected by GoI for the assessment of the sustain- ability of NGP (through the sampling method described 1.4.3 Instruments for data collection on processes earlier) was adopted on an ‘as-is’ basis, as the sample for the For information on outcomes, the household survey data analysis of the quality of service delivery processes at the set was sourced from GoI’s assessment of the sustainability district level in this study. of NGP. To assess the quality of processes by which districts implemented TSC, two data collection instruments were Selection of NGP villages. In each district, village visits were devised: an Interview Guide and a Process Rating Scale undertaken to triangulate the information collected on (Table 4). processes at the district level. Two NGP villages were visited in each district, one from the pre-2008 time group and one Interview Guide. To ensure consistency in the findings from 2008. NGP villages were selected randomly from the of the assessment, an Interview Guide was developed to TABLE 3: TYPES OF STAKEHOLDERS AND THE NUMBER OF INTERVIEWS AT THE DISTRICT LEVEL Stakeholder/Agency Whom to Interview Number of Interviews District Water and Sanitation • District Collector/Magistrate/Chief Executive 1 per district Mission/Committee Officer* • District Nodal Officer* 1 per district • District TSC Cell*members (at least one, if 1 per district available) 1 per block • Optional—selected Block Development Officers (BDOs) Gram Panchayat (GP, village • Pradhan*/elected head of the GP 1 per GP level local government) • GP members* At least 1 per GP • Households Few households could be visited at the village level to triangulate any information collected at higher administrative levels • Community members active in TSC As per context • NGO/CBO active in TSC As per context * See glossary for the meanings of these official designations. NGO: non-governmental organization; CBO: community based organization. www.wsp.org 17 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction BOX 5. STUDY SAMPLE Type of Data Source # of # of # of NGP # of Date of Data Collection States Districts Winning Households GPs District level GoI assessment Data collected in 2010, usage outcome of NGP 664 9,960 data set made available data sustainability in January 2011 District level GoI and WSP 12 56 Interviews conducted process data assessment of 112 Not in 2011 (March through service delivery applicable July) processes (this study) conduct semi-structured key stakeholder interviews. The their considerable knowledge of the relative importance of guide comprised questions on three components or stages these processes to scaling up and sustaining outcomes. The of service delivery that are considered essential for scaling final rating scale was reviewed and cleared by GoI before up and sustaining rural sanitation, namely, Catalyzing, being applied in this assessment. Implementing, and Sustaining (Annex 2).The questions were framed in such a way that an investigator could probe This is how the scoring methodology worked: in the scale, on the indicators under each process of the rating scale. each process is reflected by five indicators and each indicator is given a mark between a minimum of zero and a maximum Process Rating Scale. The Process Rating Scale was devised to of two or three marks. The allocation of maximum marks is provide a quantitative scale to interpret the findings of all such that the Sustaining component processes score higher the interviews carried out within one district. Building on than those related to Catalyzing or Implementing. The sum the scale introduced in the first round; the scale was refined total of the marks assigned to different indicators adds up to by adding and regrouping indicators to clearly identify a maximum of 100 and a minimum of 0, divided between areas of improvement and bottlenecks that hamper prog- the three components as follows—30 marks for Catalyzing, ress in service delivery. Whereas in the scale used in the first 30 marks for Implementing, and 40 marks for Sustaining. round each indicator was given equal weight, the scale in The cumulative score on the rating scale is converted into a this study assigned differential weights to indicators reflect- percentage. The scale is given in Table 5. ing a judgment on the relative importance of underlying processes to scaling up and sustaining outcomes. The differ- 1.4.4 Selection and training of interviewers ential weights assigned to each indicator were based on the Interviews for this assessment were conducted by members feedback received on the rating scale used in the first round of WSP India’s rural sanitation team which has consider- of this assessment from policy makers, practitioners, and able experience in the rural sanitation sector in India and is academics. The final rating scale for this second round was familiar with GoI’s administrative structure. All interview- thoroughly reviewed by a panel of sector experts based on ers, except one, had also participated in the first round of TABLE 4: INSTRUMENTS FOR DATA COLLECTION ON PROCESSES What does the Study Assess? How is it Assessed? Instrument Quality of service delivery processes Qualitative: Key person interviews with key program Interview Guide implementers from district to village level Quantitative: Quality of processes scored using a Process Rating Scale standard rating scale 18 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction TABLE 5: PROCESS RATING SCALE Process   Indicators Max. Score Given % Score (min=0) I. CATALYZING Strategy 1 TSC guidelines are understood and implemented by the core group 1   2 Strong political and administrative will exists to implement at 2 different levels 3 TSC implementation is being undertaken by related departments 2 4 Common understanding of implementation strategy at different 2 levels from district to GP 5 Strategy to implement TSC emphasizes collective behavior change 3 outcomes   Sub-total 10 0 0 13 Institutions 1 Nodal agency is functional and effective 1   2 Coordination mechanisms have been set up at different levels and 1 between different actors 3 Village-level institutions take the lead in implementing at the 2 community level 4 A dedicated unit for TSC with adequate staff exists at the district 3 level and is effective 5 Adequate staff and capacity exists at the block and sub-block 3 levels (e.g., cluster, GP, habitation) for implementing the program effectively   Sub-total 10 0 0 Financing 1 Additional installments are asked for on time 1   2 There are no funding bottlenecks at the district and sub-district 1 levels (funds under the control of the district)   3 Funding is used to address all components of total sanitation 2 4 Funding is used for both short-term hardware target achievement 3 and long-term sustainability (quality, usage) 5 Households invest own resources in toilet construction 3   Sub-total 10 0 0   Catalyzing component sub-total 30 0 0 2. IMPLEMENTING Demand 1 Behavior Change Communication (BCC) messages includes 2 personal hygiene and a clean environment in addition to household/   institutional sanitation 2 Mass media is used to the optimal level 2 3 BCC focuses on collective behavior change outcomes 3 4 BCC is not seen as a one-time activity and continues to focus on 3 sustainability of outcomes 5 Motivators are used to the optimal level 3   Sub-total 13 0 0 13 At state level, the rural sanitation portfolio is held by either RDD or PHED. www.wsp.org 19 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction TABLE 5 CONTINUED Process   Indicators Max. Score Given % Score (min=0) Supply 1 Multiple technology options are promoted 2   2 Technology choices respond to community preferences and are 1 affordable 3 Technology choices promoted and adopted are safe/improved 1 4 Products and services sourced are easily available 1 5 Trained masons are available for construction 1   Sub-total 6 0 0 Scale 1 TSC is implemented as a district-wide campaign in phases 2   2 Motivators are provided logistic and other support to reach all parts 2 of a district   3 Supply chain extends to hard-to-reach, difficult terrains 2 5 Strategy includes moving from ODF to ODF+ behaviors such as 2 adoption of improved hygiene behaviors like hand-washing with soap, safe disposal of garbage and wastewater, etc. 4 ODF goal focuses on households and institutions 3   Sub-total 11 0 0   Implementing component sub-total 30 0 0 3. SUSTAINING Subsidy 1 Subsidy is given as a post-construction incentive after usage is 3 verified   2 Subsidy is not treated as a unit cost to which type of toilet model is 3 fitted14 3 Subsidy is given to households that would not be able to afford the 3 cost of toilet construction without this assistance15 4 Subsidy is not given to private contractors, NGOs, etc., to build 3 toilets 5 Subsidy is given to GP/community as an incentive to reward 3 achievement of collective ODF outcomes   Sub-total 15 0 0 Monitoring 1 Monitoring results are fed back to lower levels; used in planning and 2 management   2 Monitoring systems are available at the village, block, and district 3 levels to track usage of APL, BPL, public, and institutional toilets 3 Systems are in place to ensure quality of data reported 3 4 NGP applications are thoroughly checked before being forwarded to 3 higher levels 5 Monitoring of NGP/ODF villages is undertaken regularly 3   Sub-total 14 0 0 14 According to TSC guidelines, a subsidy is intended as an incentive which may be given to BPL families if this is considered necessary for the complete involvement of the com- munity. The BPL family should undertake construction of the toilet on its own and a cash incentive may be given to it in recognition of its achievement. This indicator sought to verify whether the subsidy was, in fact, delivered as an incentive or whether it was diverted as an upfront unit cost to fund toilet construction for BPL households. 15 This indicator looked at potential errors of inclusion (households other than BPL households receive subsidy) and exclusion (BPL households eligible for the subsidy are left out) in targeting subsidy. 20 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction TABLE 5 CONTINUED Process   Indicators Max. Score Given % Score (min=0) Rewards and 1 Incentives exist for motivators and GP Pradhans which are linked to 2 Recognition achieving ODF status 2 Incentives exist for block- and district-level officials to prioritize TSC 2 outcome achievements 3 Incentives exist to sustain NGP status post-award 2   4 Incentives exist to move to ODF+ activities (e.g., safe disposal of 2 garbage and wastewater, adoption of improved hygiene behaviors   such as hand-washing with soap) that are introduced once ODF status is achieved 5 Competition-based annual rewards for cleanest GP have been 3 introduced   Sub-total 11 0 0 Sustaining component sub-total 40 0 0     TOTAL 100 0 0 this assessment which visited 22 districts in 21 states. on toilet usage was available through GoI’s study of Districts were assigned to different team members on the NGP sustainability. To use the data set of this study basis of their fluency in the local language of the state in to assess service delivery processes, its sample had to which the district was located. be replicated on an ‘as-is’ basis. Hence, the sample size and sampling methodology were not under the Before initiating the assessment, a training workshop was team’s control; held in New Delhi to familiarize team members with the • Linked with the above limitation, the districts select- revisions in the rating scale and research protocol used in ed by GoI for the study and subsequently for this the first round of this assessment. Different scenarios that assessment, are the best performers in their respec- could be encountered during district visits were discussed tive states. However, this is a partial limitation as including guidance on how to score these. Once the district states are selected from three different categories— visits started, each rating scale and district report was thor- high, average, and low—based on their performance oughly reviewed to ensure that the narrative matched the on TSC. Hence, the sample comprises ‘best districts scores given. of the best achiever states’, ‘best districts of average states’, and ‘best districts of low achiever states;’ 1.4.5 Analysis: linking processes with outcomes • Another limitation of the study relates to the fact The performance of each district in terms of outcomes that TSC processes were rated on their status as on (percentage of sample population—of NGP-winning the data collection/interview dates (in 2011), while villages/local government areas—using toilets and safely the outcome data were collected during a study disposing child feces) and the district score on the process conducted in the previous year (in 2010). While rating scale were correlated to understand if these factors are outcomes would be a cumulative result of process- linked and to what extent. es up to the previous year, the process ratings were given on the status as on 2011; therefore the findings 1.5 Limitations have to be looked at keeping this practical difficulty Some of the limitations of this study are: in view; and • GoI’s routine monitoring system for the rural sanita- • Although this study has its limitations, it contrib- tion sector in India does not track usage of toilets. utes to a growing body of work on service deliv- The most recent and reliable source of information ery processes (Box 6). Other studies have been www.wsp.org 21 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Introduction developing methodologies for assessment of process- delivery of rural sanitation programs but also for a es and a combination of efforts should help in devel- much better understanding of ‘what works.’ oping not only better tools to ‘measure’ processes for BOX 6. EXAMPLES OF OTHER STUDIES THAT ASSESS SERVICE DELIVERY PROCESSES • The African Ministers Council on Water commissioned a first round of Country Status Overviews (CSOs) on water supply and sanitation in 2006 followed by a second round in 2009-10. The objective of the CSOs is to shed light on the factors that underpin progress in the sector and what member governments can do to accelerate progress. For each country (16 included in the first round and 32 in the second round), the CSOs explore the links between inputs (financing) and outcomes (service coverage) using a ‘scorecard’ to identify areas of progress and improvements in each sub-sector; • Since 2007, WSP has been undertaking Enabling Environment Assessments (EEAs) in program countries that have ‘Scaling up Rural Sanitation’ as a core business line. These assessments analyze the programmatic and institutional conditions needed to scale up, sustain, and replicate the total sanitation and sanitation marketing programmatic and service delivery approaches. EEAs are undertaken at two points in time: a baseline assessment at the start of WSP’s intervention in a country/province and an end line assessment at the end of the intervention; and • A study was undertaken by WSP Indonesia in 2010 to identify factors associated with achieving and sustaining behavior change by communities to become ODF in East Java. This showed that communities that achieved ODF status within two months of triggering achieved markedly higher access gains. Factors associated with ‘quickly’ ODF communities include high social capital, high-quality Community Led Total Sanitation (CLTS) triggering, access to latrine supplies, easy payment terms, absence of external subsidy packages to a few households out of all, and regular monitoring. 22 Global Scaling Up Sanitation Project KEY POINTS II. Correlating Service Delivery Processes with Outcomes • Sample districts tend to score the highest on process- es under the Catalyzing component, followed by Implementing and Sustaining. This implies that while catalyzing conditions are in place, it has been challeng- ing to translate these into downstream service delivery and for sustaining outcomes; • The quality of processes adopted by districts to imple- ment TSC has a significant and positive correlation with outcomes at the district level; and • Processes under the Sustaining component, followed by Catalyzing component, seem to be more correlated with usage, compared with Implementing. Overall, it is not individual processes that determine program outcomes in a district but the sum of parts. This chapter presents the results of correlations between on each component of the process rating scale, color coded district scores on the process rating scale and district perfor- to enable identification of areas of strength and weakness. mance in terms of outcomes as measured by the percent- age population using toilets including safe disposal of child feces. Looking at individual processes in the rating scale, we find that districts scored the most in putting in place processes 2.1 How have Sample Districts Scored on to strengthen supply and financing. Since the rural sanita- Quality of Processes? tion program leverages the existing institutional structure Overall, district scores on the quality of processes tended to of the district administration, districts tend to score well be the strongest on Catalyzing, followed by Implementing on this indicator. The district process rating score is average and Sustaining components. This means that the districts on all other indicators except for rewards and incentives on scored well in terms of putting in place policies, institutions, which the average district score is the lowest of all process- and budgets related to program implementation. However, es tracked. Figure 5 presents district scores on individual it was challenging to translate these catalyzing conditions processes, color coded to identify areas of strength and into scaling up demand and supply to reach and sustain weakness (green denotes a score of more than 65, yellow is outcomes. Figure 4 shows the average scores of 56 districts more than 32, and red is less than 31). FIGURE 4: HOW HAVE SAMPLE DISTRICTS SCORED ON QUALITY OF PROCESSES IN TERMS OF COMPONENTS? (N=56 DISTRICTS) 100 80 65 59 56 60 46 % 40 20 0 Catalyzing Implementing Sustaining District Level TSC Processes Process Components www.wsp.org 23 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Correlating Service Delivery Processes with Outcomes FIGURE 5: HOW HAVE SAMPLE DISTRICTS SCORED ON QUALITY OF PROCESSES? (N=56 DISTRICTS) 100 Catalyzing Implementing Sustaining 80 72 65 68 63 61 60 60 52 44 40 31 20 0 Strategy Institution Financing Demand Creation Supply Chain & Technology Scaling Up Subsidy Monitoring Rewards & Recognition FIGURE 6: CORRELATING QUALITY OF PROCESSES WITH 2.2 Does Quality of SUSTAINING NIRMAL STATUS Service Delivery Processes Matter? District scores on the quality of processes were correlated with district performances in terms of the percentage of 100 sample population using toilets (including safe disposal of child feces) as found by GoI’s assessment of NGP sustain- ability. It was found that the quality of processes adopted 80 by districts to implement TSC had a significant and posi- Average Usage % tive correlation with usage of toilets in NGP winners. In 60 other words, the higher the quality of the processes adopted, Usage% the stronger the likelihood that NGP winners will sustain Average Process Rating % behavior change linked to toilet usage in a district (Figure 6). 40 District ratings on the three individual components of the process rating scale—Catalyzing, Implementing, and Sustaining—were also found to have significant positive 20 correlations with the district-level usage outcomes (Pearson’s r values of .642, .503, and .667 respectively, all significant 0 at the level of p<.01) (Figure 7). These correlations indicate 0 20 40 60 80 100 that: Process Rating % • The higher the quality of processes adopted by a Strong (r=+.634, p<.01) district, the greater the likelihood of sustaining State (Each point = a District) usage/behavior change among the population in Andhra Pradesh Himachal Pradesh Rajasthan NGP-winning GPs; and Bihar Karnataka Tripura Chhattisgarh Kerala Uttar Pradesh • Sustaining, followed by Catalyzing, seems to be more Haryana Maharashtra West Bengal Fit line for Total correlated with usage compared with the processes in the Implementing component. 24 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Correlating Service Delivery Processes with Outcomes FIGURE 7: CORRELATING SAMPLE DISTRICT SCORES ON PROCESS COMPONENTS WITH DISTRICT PERFORMANCE IN TERMS OF TOILET USAGE Catalyzing Sustaining 100 100 80 80 Average Usage % Average Usage % Usage % 60 60 Usage% Average Process Rating % Average Process Rating % 40 40 20 20 0 0 0 20 40 60 80 100 0 20 40 60 80 100 Catalyzing Rating % Sustaining Rating % Moderate, Pearson’s r=+.642, p<.01 R2 Linear = 0.412 Moderate, Pearson’s r=+.667, p<.01 R2 Linear = 0.445 State (Each point = a District) State (Each point = a District) Andhra Pradesh Himachal Pradesh Rajasthan Andhra Pradesh Himachal Pradesh Rajasthan Bihar Karnataka Tripura Bihar Karnataka Tripura Chhattisgarh Kerala Uttar Pradesh Chhattisgarh Kerala Uttar Pradesh Haryana Maharashtra West Bengal Haryana Maharashtra West Bengal Fit line for Total Fit line for Total Implementing 100 80 Average Usage % 60 Usage% Average Process Rating % 40 20 0 0 20 40 60 80 100 Implementing Rating % Moderate, Pearson’s r=+.503, p<.01 R2 Linear = 0.253 State (Each point = a District) Andhra Pradesh Himachal Pradesh Rajasthan Bihar Karnataka Tripura Chhattisgarh Kerala Uttar Pradesh Haryana Maharashtra West Bengal Fit line for Total www.wsp.org 25 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Executive Summary Individual processes were also correlated with usage found that although it was possible to identify high, aver- outcomes (Table 6). While a majority of the processes had age, and low scorers based on total rating scores, within a significant and strong positive correlation (r>0.5 or 0.6), these scoring groupings, high scorers did not score well three processes (demand creation, scaling up, and rewards) across the board on all processes measured through the out of the nine processes were found to have a moder- rating scale. Similarly, poor scorers also did not score poor- ate to weak correlation at the same level of significance ly across all processes that were measured and tended to (p< 0.01), as shown in Table 6. Since this assessment evalu- match and, in a few cases, exceed average or high scorers ates the service delivery system at the district level, the sum (Results in the next section). Hence, it is not individual of parts rather than standalone processes are more impor- processes that determine program outcomes in districts but tant. This is also reflected in district scores wherein it was the sum of their parts. TABLE 6: RESULTS OF INDIVIDUAL PROCESS CORRELATION WITH USAGE OUTCOMES (N=56) Component Average Component Process Average Process Process Component Correlation Score as per Correlation Process Rating Rating Scale Catalyzing 65 .642** Strategy 63 .636** Institutions 65 .617** Financing 68 .535** Implementing 59 .503** Demand 52 .455** Supply 72 .615** Scale 61 .445** Sustaining 46 .667** Subsidy Delivery 60 .645** Monitoring 44 .660** Rewards 31 .432** **All correlations (Pearson’s r values) are positive and significant at p<.01. 26 Global Scaling Up Sanitation Project III. KEY POINTS What Explains the • The largest gaps between high and low scoring Differences in District districts were found in the following processes: strat- egy in the Catalyzing component, demand creation in Scores on Service the Implementing component, and monitoring in the Delivery Processes? Sustaining component; and • Among the nine processes rated, districts scored the highest on processes related with supply and lowest on processes related with rewards and recognition, followed by monitoring. In this chapter, sample districts are divided into three cate- of the overall process rating score. Figure 8 shows where gories—high scorers, average scorers, and low scorers— the largest gaps are found between high and low scorers based on their scores on the process rating scale. It provides on each process in each component of the rating scale, a detailed analysis by component and process of the simi- namely in strategy under Catalyzing, demand creation larities and differences in terms of processes adopted by under Implementing, and monitoring under the districts on different points on the performance curve in Sustaining component. terms of the quality of the processes. A detailed analysis of the differences and similarities In order to identify what differentiates district scores between high, average, and low scoring districts is also in terms of the quality of service delivery processes, the presented by process under each component for all the sample of 56 districts was distributed into three categories nine processes scored through the rating scale. —high, average, and low scorers—in descending order FIGURE 8: 56 DISTRICTS’ AVERAGE SCORES ON QUALITY OF PROCESSES (N=56 DISTRICTS) Catalyzing Implementing Sustaining 100 80 60 40 20 0 Strategy Institutions Financing Demand Supply Scale Subsidy Monitoring Rewards High Scorer Average Scorer Low Scorer Arrows indicate largest gaps between high and low scorers www.wsp.org 27 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Differences in District Scores on Service Delivery Processes 3.1 Catalyzing Program FIGURE 9: COMPARING DISTRICT SCORES: STRATEGY (N=56) Implementation Strategy. Figure 9 shows sample 100 districts’ scores on strategy, divided High scorers tend to prioritize by high, average, and low scorers. 80 rural sanitation In terms of the strategy adopted to implement TSC, high scorers show 60 a common understanding of the Low scorers tend to focus on the TSC strategy whereas in low scor- toilet construc- tion target rather ers, typically, a disconnect was found 40 than behavior between the paper strategy approved change at higher policy levels that emphasizes 20 outcomes and its diluted understand- ing at the implementation level. Rural 0 sanitation is often not a priority, espe- TSC Strong Implemen- Common Strategy guidelines political tation under- emphasizes cially when compared with more high understood and undertaken standing of collective Strategy and by related behavior profile sectors such as water supply. imple- administra- depart- implemen- change tive will tation However, high scorers have succeeded mented by exists at ments strategy at core group different different in making sanitation a priority at both levels levels administrative and political levels. High Scorer Average Scorer Low Scorer High scorers also tend to put in place Arrows indicate largest gaps between high and low scorers a strategy that emphasizes collective behavior change. By contrast, as can be seen from Figure 10, this is the weakest link in the strategy of average FIGURE 10: COMPARING DISTRICT SCORES: INSTITUTIONS (N=56) to low scorers which tend to empha- size outputs such as toilet construc- 100 tion rather than behavior change. An Although there is a provision to get staff area of improvement across districts is on deputation/ 80 coordination between related depart- contract appointment, many ments. Findings show that despite positions were found vacant and 60 sanitation having close links with capacity is particularly weak at sectors such as nutrition, health, and the sub-district level local government, there is a tendency 40 to see it as a preserve of the nodal agency (either RDD or PHED) 20 appointed to implement the rural sanitation program. 0 Nodal Coordination Village level Dedicated Adequate staff Institution agency is mechanisms institutions District and capacity Institutions and Capacity. In terms of functional are take Sanitation exists at block and effective the lead at Cell and sub-block the institutional framework, the nodal effective community exists and is levels level agency was found to be functional effective in most sample districts (Figure 10). High Scorer Average Scorer Low Scorer Village-level institutions were also Arrows indicate largest gaps between high and low scorers found to play a key role reflecting the 28 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Differences in District Scores on Service Delivery Processes decentralized design of the program. FIGURE 11: COMPARING DISTRICT SCORES: FINANCING (N=56) However, coordination mechanisms, though found to be in existence, 100 were not effective in a majority of Outcomes depend on how districts. A key challenge for districts 80 funding is utilized and not only on that are weak scorers is that despite the quantum of sanctioned posts many key staff posi- funds allocated 60 tions (for example, district coordi- nator, functional specialists in areas such as monitoring communications, 40 and so on) were found vacant. Box 7 describes the institutional structure 20 for Alappuzha, a Nirmal district in Kerala, which shows how a high scor- 0 er has put in place an institutional Additional No funding Funding is Funding Households Financing instalments bottlenecks used to used for invest own model to scale up TSC. are asked at district address all both resources in for on time and components short-term toilet sub-district of total and construction sanitation long-term Financing. As can be seen from Figure levels goals 11, it was reported across sample High Scorer Average Scorer Low Scorer districts that there were no delays Arrows indicate largest gaps between high and low scorers or bottlenecks in funding, reflecting that the rural sanitation allocation was released by GoI through state governments for utilization as per the BOX 7. INSTITUTIONAL MODEL TO SCALE UP TOTAL SANITATION: demands from the districts. However, EXAMPLE OF A DISTRICT THAT HAS A HIGH SCORE ON QUALITY OF once the funds were released, high PROCESSES scorers tended to use the funding to Alappuzha district in Kerala comprises 73 GPs and 12 blocks. All GPs and address all components of total sani- Block Panchayats in this district won NGPs during 2007-09 and the district tation including household sanita- was declared Nirmal in 2009. Alappuzha is the smallest district in Kerala in tion, school sanitation, and waste terms of area but has the highest density of population (1,492 persons/square management. High scorers also tend- kilometer). Sanitation has always been a challenge in this district due to the ed to focus not only on achieving presence of high water table areas and a coastal belt. but also sustaining behavior change to end open defecation. Under TSC, The District Sanitation Samiti (DSS) attached to the District Panchayat (DP) is despite the availability of an incentive responsible for the implementation of TSC in the district. While DSS is chaired for toilet construction, high scorers by the President of the DP, the District Collector (DC) is the Executive Director tended to focus on mobilizing users to of DSS. DSS meets regularly to review progress of the work. Due to frequent change their sanitation situation and outbreaks of sanitation linked diseases such as chikungunya (a viral illness invest their own resources in toilet spread by bites of infected mosquitoes) and malaria, this is a priority sector construction. Such an approach was among district administrators. found to be more sustainable than one which provides toilets as ‘gifts-in- DSS has a District Coordinator and two Assistant Coordinators (all on depu- aid’ (Tremolet 2010). tation from RDD) and one Data Entry Operator (on contract appointment). District and Block Resource Teams have been set up in the district to help in www.wsp.org 29 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Differences in District Scores on Service Delivery Processes also sustain it. By contrast, weak scor- the implementation of TSC. At the GP level, the Village Extension Officer (VEO) ers tended to treat BCC as a one-time is responsible for the implementation of TSC. The program is implemented activity and did not invest adequately through the GP. A Health Promotion Team has been set up at the ward level in making inter-personal and mass in the GPs mainly with volunteers from Kudumbasree (Kerala Government’s communication effective (Box 8). women’s empowerment and poverty eradication program) Self-Help Groups (SHGs), health department staff, and so on. Supply and Technology Promotion. The TSC Guidelines advocate The GP President and ward members provide leadership in identifying benefi- informed technology choices and ciaries and motivating them at the village level. There is no NGO support in the setting up of alternate supply chan- implementation of TSC in the district. The VEO provides monitoring support nels such as RSMs to facilitate supply and also supervises construction work. Beneficiaries are responsible for the of products and services, if required. construction of their own toilets and BPL households receive monetary incen- In the sample, districts across scor- tives from TSC on completion of the work. The beneficiaries usually spend ing bands tended to promote safe much more than the incentive they receive as they tend to construct pucca or improved technology under the (permanent) toilets with good superstructures and, in some cases, attached program and did not report any bathrooms. The school and anganwadi sanitation program is implemented difficulties in the availability of sani- by Panchayati Raj Institutions (PRIs) which are responsible for managing tary products or masons (Figure the schools in the state as per the decentralization program. At the GP level, 13). Whereas high scorers tended to Sanitation Committees exist at the ward level under the leadership of ward promote informed choice and multi- members of the GP. These committees also help in the implementation of TSC, ple technology options, low scorers especially the waste management program. were found to scale up construction of a single technology model regardless 3.2 Implementing FIGURE 12: COMPARING DISTRICT SCORES: DEMAND (N=56) Program Goals Demand Creation. TSC guide- 100 lines advocate a demand-driven Focus on behavior change approach to rural sanitation and up 80 rather than toilet construction is to 15 percent of a district’s budget for linked with rural sanitation is allocated for IEC. outcomes 60 Districts have the flexibility to imple- ment a demand-driven approach 40 and utilize IEC funds based on their context and capacity. As shown in Figure 12, sample districts were 20 almost at par as far as the develop- ment of communication material 0 was concerned. However, high scor- BCC Mass BCC BCC is not Motivators Financing messaging media is focuses on a one-time are used to ers tended to diverge after this point goes used to the collective activity the optimal beyond optimal behavior level as they were able to exploit available household/ level change channels more effectively to ensure institutional sanitation that the messages reached the target High Scorer Average Scorer Low Scorer audience over a period of time to not Arrows indicate largest gaps between high and low scorers only motivate behavior change but 30 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Differences in District Scores on Service Delivery Processes of whether it reflected household FIGURE 13: COMPARING DISTRICT SCORE: SUPPLY AND TECHNOLOGY choice or affordability (Box 9). PROMOTION (N=56) 100 Scaling up. On average, sample districts did not report any prob- lems in terms of accessing sanitary 80 products, even in difficult terrains, Low scorer's as can be seen from the consolidated tendency is to 60 promote a single findings in Figure 14. In every other technology rather than a menu of respect, however, high scorers tend- options 40 ed to adopt an approach that was amenable to working across commu- nities rather than creating pockets of 20 excellence. Firstly, high scorers tended to implement the program in phases 0 Multiple Technology Technology Sanitary Trained Supply technology choices choices products masons are options reflect promoted/ and services available for BOX 8. PROMOTING promoted consumer adopted are easily construction preferences are safe/ available SANITATION AS TOILET and improved CONSTRUCTION, NOT affordability BEHAVIOR CHANGE High Scorer Average Scorer Low Scorer Arrows indicate largest gaps between high and low scorers In a low scoring sample district, under the IEC component, NGOs have been engaged on contract FIGURE 14: COMPARING DISTRICT SCORE: SCALING UP (N=56) to undertake activities such as wall paintings, film shows, street 100 plays, distribution of pamphlets, Spread too thin? Low scorers tend and so on. Door-to-door visits to tackle as many 80 are also undertaken by village GPs as possible at one go rather functionaries such as anganwa- than implement in phases 60 di workers, health workers, and school teachers. However, the content of the messaging tends 40 to focus on individual household toilet construction rather than 20 the need for behavior change to achieve ODF status. The district 0 has used ‘achievement of NGP’ TSC Motivators Supply chain Strategy ODF goal Scaling Up implemen- provided extends to includes focuses on as a key trigger with GPs and, ted as a logistic and hard-to- moving from households district-wide other reach, ODF to and subsequently, the focus has been institutions campaign in support to difficult ODF+ on completing toilet construction phases reach all terrains parts of a before the NGP verification team district arrives rather than motivating High Scorer Average Scorer Low Scorer behavior change. Arrows indicate largest gaps between high and low scorers www.wsp.org 31 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Differences in District Scores on Service Delivery Processes BOX 9. WHOSE TOILET? LACK OF HOUSEHOLD INVOLVEMENT IN SELECTING TOILET TECHNOLOGY The TSC Guidelines state that an incentive can be given to a household to recognize its achievement of stopping open defecation. However, in practice, this amount is sometimes treated as a subsidy to construct a toilet rather than as an incen- tive to reward behavior change. In one of the low scoring sample districts, a single offset leach pit, with slab and rural (high gradient) pan, is the model promoted for all households regardless of preference or geo-hydrological conditions. In addition to the incentive amount available under the program for BPL families, the district has made incentives available to APL families as well under the state policy. This ‘incentive’ amount is treated as a unit cost or subsidy for constructing a household toilet. Accordingly, the toilet model promoted by the district is fit to the incentive/subsidy amount. This comprises a sub-structure and four walls of height up to 4 feet for both BPL and APL families, and every family is expected to install a door and a roof. The district has extended financial support for five RSMs but none of these is functional. Even so, there are no difficul- ties reported in accessing sanitary products due to the existence of a well-established private market for most materials except for rural pans. Subsequently, the GP purchases all materials required for toilet construction locally; rural pans are procured by the district and made available through the sub-divisional office. The GP engages local masons and monitors the construction. Households play a minimal role in selecting the toilet model, engaging masons, and construction, and it was found that most did not invest in installing a door and a roof to complete the toilet sub-structure provided or, in fact, use the toilet constructed for them. BOX 10. PHASED APPROACH TO TSC IMPLEMENTATION IN RAIGAD DISTRICT, MAHARASHTRA In Raigad, a high scoring district, phased take-up of GPs is based on the following three criteria: 1) relatively small GPs in terms of population; 2) at least a third of the households use toilets; and 3) positive outlook of elected representatives towards the rural sanitation program. Training is targeted at village health extension workers and elected representatives. A parallel initiative is the Swacchata Dhoot (sanitation ambassador) concept targeted at school teachers and selected students. The district follows a post-construction incentive system for households. The GP, in coordination with the school, is responsible for promoting safe sanitation through a community-led approach for achieving 100 percent ODF status. Utilizing other platforms such as the National Service Scheme in colleges, organizing exposure visits to GPs that have won the state sanitation prize (Sant Gadge Baba Gram Swachata Abhiyan award), training programs for masons, and organizing the Nirmal Raigad Neta Rath (Clean Raigad Leaders Rally) are allied initiatives. A strong political will to focus on sanitation is reported. The role of some members of the Zila Panchayat (ZP) in encouraging and ensuring support for rural sanitation is reported. rather than spreading efforts too thinly by implementing environment and, moving beyond that, to waste manage- across all local governments at once (Box 10). Secondly, ment. trained motivators were provided with logistic and other support to enable them to reach communities in different 3.3 Sustaining Outcomes parts of a district. Last, but not the least, high scorers tend- Subsidy Delivery. Processes linked with subsidy delivery ed to be inclusive in that they interpreted the program goals have a high correlation with outcomes achieved at the not just in terms of household sanitation but as an ODF district level. Across sample districts, as can be seen from 32 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Differences in District Scores on Service Delivery Processes Figure 15, subsidy was usually not FIGURE 15: COMPARING DISTRICT SCORE: SUBSIDY DELIVERY (N=56 ) given to a private contractor or 100 external agency such as an NGO to High scorers treat subsidy as construct toilets. Beyond that, high incentive for behavior change 80 scorers tended to follow the program whereas low guidelines in letter and in spirit. scorers treat subsidy as unit For instance, high scorers released cost released 60 after construction the subsidy as a post-construction incentive rather than as an upfront 40 payment. Secondly, they focused on mobilizing the community and 20 motivating households to invest their own resources in toilet construction. Hence, the subsidy was not treated 0 Subsidy is Subsidy is There are Subsidy is Subsidy is Subsidy Delivery as a unit cost to determine which given as a not treated minimal not given to given to post- as a unit errors of private GP/commu- model of toilet technology should construction cost to targeting contractor, nity as an be constructed (Box 11).The practice incentive which type NGO to incentive for after usage of toilet build toilets collective of treating the subsidy as an incen- is verified model is ODF fitted outcome tive not just for individual household High Average Low toilet construction but for achieve- Performers Performers Performers ment of community-wide ODF Arrows indicate largest gaps between high and low scorers status was found in very few sample districts. FIGURE 16: COMPARING DISTRICT SCORE: MONITORING (N=56) Monitoring. Processes related to 100 monitoring had the highest correla- Checking NGP applications to tion with outcomes among the nine see if they meet 80 processes scored through the rating the minimum eligibility criteria scale. Monitoring is also an example for the prize would improve 60 where high scorers are clearly distin- success rate of guished from average and low scorers. winners to applicants As can be seen from Figure 16, high 40 scorers tended to use monitoring not just for reporting to higher levels but 20 also to establish systems to ensure that monitoring results were used as a way 0 for improving planning and manage- Monitoring Monitoring Systems are NGP NGP/ODF results are in place to applications villages ment of the program. As TSC provides systems are Monitoring fed back to available at ensure are monitored incentives for the construction of BPL lower levels; different quality of thoroughly regularly used in levels to data checked toilets, the monitoring in average and planning and track usage reported before manage- nomination low scorers tended to focus on achieve- ment of APL, BPL, public and ment of targets linked with BPL toilet institutional toilets construction. However, high scor- High Average Low ers tended to ensure that monitor- Performers Performers Performers ing covered different segments of the Arrows indicate largest gaps between high and low scorers www.wsp.org 33 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Differences in District Scores on Service Delivery Processes BOX 11. SUBSIDY AS UNIT COST FOR SELECTING TOILET MODEL The incentive available under the national rural sanitation campaign is supposed to be a reward for a family that stops open defecation and switches to using a toilet. It is not intended as a unit cost to cover the cost of constructing a toilet. In fact, a household can construct a toilet of its choice based on affordability and preference. However, since monitoring of outcomes is weak, the use of incentives as unit costs for constructing toilets persists. In some sample districts that score low on the quality of processes and report weak outcomes, subsidy plays a key role in determining the choice of technology promoted. When the incentive/subsidy was ` 600 per toilet, districts instructed GPs to construct a direct single leach pit toilet fitted with a pan and trap and no superstructure. When the incentive/subsidy increased to ` 1,500, the model was modified to make the leach pit off-set from the slab. When the incentive/subsidy was again increased to ` 2,500, the structure was modified to include a junction box with provisions for an additional pit and four walls raised to a height of up to 4 feet. community and not just BPL house- goals; this has not yet translated into not always monetary but often focus holds. Most importantly, high scor- a high correlation with outcomes on recognition of staff that performs ers tended to have systems in place to (Figure 17). It is also important to well (Box 13). ensure the quality of information being note that incentives offered by high reported (Box 12). This is reflected not scorers to program implementers are just in routine monitoring but also in the checking of NGP applications FIGURE 17: COMPARING DISTRICT SCORE: REWARDS (N=56) before nominating local governments for the award. And finally, as NGP is a one-time award with winners only ever 100 verified once at the time of the appli- cation, there is a tendency to ‘forget’ 80 NGP winners once they have won the award. High scorers, however, were 60 found to institute systems for regular By design, NGP monitoring of villages even after they is a one-time award and there achieved ODF status. 40 is a tendency to ‘forget’ winners once they have Rewards and Recognition. The study 20 won the prize. findings show that while incentives exist for local government leaders and 0 frontline workers, block and district Motivators Block and Incentives Incentives Competition and GP district level exist to exist to based officials were generally not targeted to Rewards pradhans officials sustain NGP move from annual prioritize rural sanitation. Although incentivized incentivized status ODF to rewards for to achieve to prioritize post-award ODF+ cleanest GP this is an area where high scorers are ODF status TSC activities outcomes clearly distinguished from average and low scorers in that high scorers High Average Low Performers Performers Performers have introduced incentives to not Arrows indicate largest gaps between high and low scorers just achieve but exceed TSC program 34 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Differences in District Scores on Service Delivery Processes BOX 12. CROSS-VERIFICATION OF TSC MONITORING DATA In one of the high scoring sample districts, it was found that TSC is monitored on a routine basis through monthly and quar- terly reviews at the district level. According to the Project Officer of the District Rural Development Agency, TSC and national livelihoods program are seen as flagship programs in the district. Over and above the parameters specified for the online monitoring of TSC such as expenditure and toilet construction against targets, the district has also initiated its own verification of toilet usage, an indicator that is currently not tracked by the national routine monitoring system. This is based on cross-verification by teams comprising block officials and members of support organizations. Support organizations are NGOs engaged by the district to support community mobilization for rural sanitation. Cross-verification means that teams from one block visit local governments in another block in such a way as to avoid reciprocal verification. In other words, Block A team visits Block B, which in turn visits Block C. All 228 GPs in the district are covered in this exercise and the findings showed that usage of toilets is quite high and incidence of open defeca- tion varies from 2 to 16 percent across GPs. BOX 13. NON-MONETARY INCENTIVES: EXAMPLE OF DAKSHIN KANNADA DISTRICT IN KARNATAKA In Dakshin Kannada district, the leadership has played a key role in motivating the staff to work on rural sanitation. No cash incentives are given but non-monetary recognition is highlighted through simple gestures such as recognizing change agents at public functions such as Republic Day celebrations, giving an opportunity to staff to lead and teach others, letters of appreciation from the district Chief Executive Officer, and an opportunity to be a part of the team that receives the NGP prize from the President of India in New Delhi. www.wsp.org 35 IV. KEY POINTS Identifying Priorities • Among the three process components, the key priority for for Sector Reforms reform is putting in place effective processes to sustain outcomes, followed by processes under Catalyzing and Implementing; and • Within Sustaining, the key priority should be to strength- en monitoring followed by subsidy delivery. Within Catalyzing, the key priority for reform is strategy and within Implementing the key priority is demand creation. The process rating scale helps to identify areas of strength in terms of an enabling strategy, institutional structure, and and weakness in service delivery processes at the district budgetary allocations as well as implementation processes. level. Table 7 shows for each component and process, the However, translating these catalytic conditions into service correlation strength (r-value) with sanitation outcomes and delivery outcomes and sustaining these outcomes has been the process rating score. Components and processes within a challenge for program implementers. components are arranged in descending order of correla- tion strength. As TSC is restructured into the Nirmal Under the Sustaining component, monitoring has the Bharat Abhiyan, the component and processes highlighted strongest correlation with outcomes but the second lowest in bold in Table 8 represent the key priorities for reform. rating score among the nine processes tracked (Table 8). The remainder of this section contains recommendations The performance of a program is often driven by what is for sector reforms, organized by components. monitored. The present monitoring system for rural sanita- tion focuses on inputs and outputs achieved in the short 4.1 Sustaining Outcomes term (toilet construction) rather than sustaining behavior As can be seen from Table 8, that among the three compo- change (toilet usage). The NGP monitoring system looks at nents analyzed, Sustaining has the strongest correlation community-wide ODF status but, since it is designed as a with usage/behavior change outcomes but the lowest rating one-time prize, the program’s focus shifts from winners to score. This indicates that districts have been relatively more those GPs that are still eligible to win. Surveys of sanitation successful in putting in place a framework for service delivery behavior in rural areas make information available in the TABLE 7: PRIORITIZING PROCESSES FOR REFORM (N=56) Average Process Correlation r-value Component Process Process Average Process Rating Score Correlation Rating 46 .667** Sustaining .660** Monitoring 44 .645** Subsidy 60 .432** Rewards 31 65 .642** Catalyzing .636** Strategy 63 .617** Institutions 65 .535** Financing 68 59 .503** Implementing .615** Supply 72 .455** Demand* 52 .445** Scale 61 *Supply has a strong correlation but the sample districts have already scored reasonably well on supply. At the same time, sample districts have scored lowest on demand in the implementing component. Hence, demand creation is highlighted as the priority for reform. **All correlations (Pearson’s r values) are positive and significant at p<.01. 36 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Identifying Priorities for Sector Reforms public domain every three to four years by when it is often utilized as an incentive rather than a unit cost. Over the 12 too late to make a mid-course correction. A mechanism is years of TSC, the amount of subsidy has been increased at needed between these extremes to ensure that timely and periodic intervals starting from ` 600 in 2005 to ` 3,500 reliable information is available to check the pulse of the in 2011. Accordingly, many districts modified the toilet program at periodic intervals. model to fit the unit cost of ` 600 and also at higher levels of the subsidy. Hence, rather than the amount of subsidy Some recommendations for reform, looking to the oppor- per se, the focus can be on ensuring that this is released tunity provided by the Nirmal Bharat Abhiyan, include: after behavior change and not after toilet construction. In • The proposal to release the NGP prize money in two a scenario where demand creation is weak, the behavior installments, with the second installment released change incentive tends to be used as a subsidy for toilet subject to Nirmal status being sustained for at least construction. In other words, behavior change needs to one or more years needs to be operationalized. come before subsidy and toilet construction rather than the Re-survey of previous years’ winners can be under- other way round. taken by survey agencies along with the current year’s applicants; While rewards were not found to be as strongly correlated • Those GPs that achieve Nirmal status may be recog- with outcomes such as monitoring and subsidy delivery, nized through a state-level prize in the first instance. this process has the lowest rating score among the nine Those that are found to sustain their Nirmal status processes measured. Since sanitation is often a poor cousin for one or more years after winning the prize may to more high profile sectors such as water supply or health, be felicitated by the President of India at a national this indicates that there is considerable scope to improve prize giving ceremony which carries more prestige; mechanisms to motivate program functionaries to focus on • Some states (Maharashtra, Himachal Pradesh, rural sanitation and also to go beyond ODF or NGP to Haryana, and Karnataka) have also introduced sustaining outcomes. annual competitions between GPs to be recognized as the cleanest and these states have performed 4.2 Catalyzing Outcomes comparatively better in terms of progress on Sample districts scored the highest on processes related with program goals. This means that GPs need to not the Catalyzing component compared to the Implementing only sustain but exceed their Nirmal status in or Sustaining component. Although strategy was found to order to compete successfully with their peers. have the strongest correlation with outcomes, this was also Such programs put the onus on local government the process with the lowest rating score within this compo- to ensure that community-wide ODF status is nent. A key weakness of the current program is that although sustained; and districts (and states) have the flexibility to develop a strategy • The TSC Management Information System (MIS) that responds to the local context and capacity, most tend makes data available online and is updated on a to follow the TSC Guidelines in letter rather than in spirit. monthly basis. This can be further strengthened by Hence, although the guidelines advocate behavior change utilizing mobile-to-web technology which has the and community-wide ODF status, delivery systems adapt potential to make information on outcomes avail- to what is measured; in this case, it is toilet construction able at shorter intervals compared to conventional and expenditure rather than behavior change. pen and paper surveys. In terms of institutions, findings from the sample districts With respect to subsidy delivery, there are proposals to show that while a nodal agency is functional in a majority increase the amount of subsidy available and also to make of the cases, its capacity is weak. There is a need to create a APL families eligible for this under the Nirmal Bharat dedicated sanitation cell at the district level which includes Abhiyan. This assessment’s findings show that there is still specialists with expertise in monitoring, communications, some way to go in terms of ensuring that the subsidy is and so on, to manage the campaign. This staff can be www.wsp.org 37 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Identifying Priorities for Sector Reforms engaged on deputation from other departments or appoint- scorers do best in (and the low or average scorers show a ed on contract. In addition, there is a need to strengthen large gap in) are: offering technological choices respon- capacity to implement the program at the sub-district levels sive to community preferences, that too in an affordable as well. way at the district level, simultaneously not treating the subsidy as the unit cost, to which a type of toilet model Sample districts scored the highest rating on financing is fitted for supply-driven construction at the village (GP) within the Catalyzing component. The lack of bottlenecks level. These large gaps are not conducive for promoting reported in accessing financing for sanitation indicates behavior change because when users are not involved in that this is not a constraint for the sector and this situa- choosing the technology, they are often reluctant to break tion is likely to continue as the Nirmal Bharat Abhiyan is the habit of open defecation. This leads to poor quality slated to have a larger allocation than TSC. The challenge and/or incomplete construction of toilets. For example, for districts is to make effective use of these resources to missing doors, reduced height of walls, and no lining/ focus on achievements of outcomes rather than short-term covers of pits since the household does not bother to targets. finish a toilet which it has not demanded in the first place. In some cases, constructed toilets are used for storage or 4.3 Implementing Outcomes bathing and washing. In fact, inadequate demand creation In terms of Implementing, the priority for reform is process- is intricately linked to this supply sub-process that seems es linked with demand creation which has the lowest score to be the indicator most correlated with usage outcomes. among processes in this component. There is a mispercep- A good demand creation contributes to the community’s tion that demand creation can come after toilet construc- voice being reflected in the choice of technology being tion. On the contrary, demand creation begins with entry adopted/supplied. This further supports the case for into a community and the process adopted to achieve sani- making demand creation a priority for future reform. tation goals at the entry. From a programmatic point of view, undertaking demand creation after toilet construction Finally, in terms of scale, a key area of improvement for is akin to putting the cart before the horse. Also, once toilet districts is ensuring prioritization of the program in terms construction targets are met and reflected in routine moni- of selection of villages and also behaviors. Achieving total toring, the incentive to ensure quality of construction and sanitation implies that several behaviors are targeted. motivate behavior change is reduced. However, not all changes can be made at once and hence the focus can start with addressing open defecation and Sample districts scored well overall on supply of sanitation then tackling other behaviors such as waste management. products and services and technology promotion as indi- Many districts also reported training of motivators for cated by the fact that this process had the highest average community mobilization. However, expecting motivators rating score of 72. While accessing sanitary products was to work on a voluntary basis is not sustainable and training not reported as a bottleneck in even relatively remote areas must be accompanied by provisions to ensure that motiva- in the sample districts, the assessment’s findings show that tors are provided with logistical and other support to work in many cases there was a tendency to promote a single across communities. Going further, motivators can also be technology model which fits the subsidy available for the provided with performance-linked incentives that are relat- construction of toilets. The sub-processes that the high ed to outcomes. 38 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation References 1. AMCOW. 2011. Pathways to Progress: Transitioning to and Sector Reform to Accelerate Access to Improved Rural Country-led Service Delivery Pathways to meet Africa’s Sanitation. Washington DC: Water and Sanitation Water Supply and Sanitation Targets. Program, The World Bank. 2. Cialdini, Robert B. 2003. ‘Crafting normative messag- 12. Robinson, A. 2010. Sanitation Financing in Rural es to protect the environment’,Current Directions in Cambodia: Review and Recommendations. Study Psychological Science 12: 105–109. supported by: Asian Development Bank and Water and 3. Government of India. 2004. Guidelines on Central Rural Sanitation Program. Sanitation Programme: Total Sanitation Campaign. 13. Sanan, D. and S.G. Moulik. 2007. An Approach that New Delhi: Department of Drinking Water Supply, Works: Community Led Total Sanitation in Rural Areas. Ministry of Rural Development. New Delhi: Water and Sanitation Program, The World 4. Government of India. 2008. Guidelines on Central Rural Bank. Sanitation Programme: Total Sanitation Campaign. 14. Spears, D. 2012. From the Total Sanitation Campaign New Delhi: Department of Drinking Water Supply, to the Nirmal Bharat Abhiyan: Risks and Opportunities. Ministry of Rural Development. Unpublished. 5. Government of India. 2011. Assessment Study of Impact 15. Srinivasan, G. 2012. Tokenism will not do. The Hindu, and Sustainability of NGP. New Delhi: Department June 14. of Drinking Water Supply, Ministry of Rural 16. TARU. 2008. Impact Assessment of Nirmal Gram Development. Puraskar Awarded Panchayats. New Delhi: The Action 6. Government of India. 2012. CCEA approves enhanced Research Unit. allocation for sanitation; Total Sanitation Campaign 17. The World Bank. 2004. India Burden of Disease rechristened as Nirmal Bharat Abhiyan. New Delhi: Statistics, available at http://genderstats.worldbank.org/ Cabinet Committee on Economic Affairs, notification hnpstats/files/BOD4.xls. Accessed in March 2010. available at www.ddws.nic.in. 18. Tremolet, S.,P. Kolsky and E. Perez. 2010. Financing 7. Office of the Registrar General and Census On-site Sanitation for the Poor: A Six Country Commissioner. 2001. Census of India. New Delhi: Comparative Review and Analysis. Washington DC: Ministry of Home Affairs, Government of India. Water and Sanitation Program, the World Bank. 8. Kar, K. 2003. Subsidy or Self Respect? Participatory Total 19. Water Aid. 2008. Feeling the pulse: a Study of the Total Community Sanitation in Bangladesh. IDS Working Sanitation Campaign in Paper 184. Brighton: Institute of Development Studies. 20. Five States. New Delhi: Water Aid India. 9. Mukherjee, N. 2011. Factors Associated with Achieving 21. WSP. 2006. Study of Best Practices in Rural Sanitation in and Sustaining Open Defecation Free Communities: India: Working towards an Open Defecation Free Rural Findings from East Java. Indonesia: Water and Sanitation India. Draft Report prepared for WSP. New Delhi: Program, the World Bank. Water and Sanitation Program, the World Bank. 10. RGNDWM. 2005. Nirmal Gram Puraskar: a National 22. WSP. 2010. A Decade of the Total Sanitation Campaign: Award under Total Sanitation Campaign. New Delhi: A Rapid Assessment of Processes and Outcomes. Vols. 1 Rajiv Gandhi National Drinking Water Mission. and 2. New Delhi: Water and Sanitation Program, The 11. Rosensweig, R., E. Perez and A. Robinson. 2012: Policy World Bank. www.wsp.org 39 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Annexes Annex 1: Selection of Household Sample by State and District States Districts Number of NGP winners selected Households @ 15 per NGP winner Pre-2008 2008 Total Andhra Pradesh Guntur 6 6 12 180 Karimnagar 6 6 12 180 Krishna 6 6 12 180 Khammam 6 6 12 180 West Godavari 6 6 12 180 Bihar Katihar 6 7 13 195 Vaishali 8 7 15 225 Chhattisgarh Rajnandgaon 7 7 14 210 Korba 7 7 14 210 Haryana Karnal 8 6 14 210 Panipat 8 6 14 210 Kurukshetra 2 6 8 120 Himachal Solan 2 7 9 135 Pradesh Mandi 12 7 19 285 Karnataka Uttara Kannada 6 6 12 180 Dakshin Kannada 6 6 12 180 Shimoga 6 6 12 180 Kerala Thiruvananthapuram 6 6 12 180 Alappuzha 6 6 12 180 Ernakulam 6 6 12 180 Thrissur 6 6 12 180 Maharashtra Solapur 6 6 12 180 Nagpur 6 6 12 180 Bhandara 6 6 12 180 Ahmednagar 6 6 12 180 Raigad 6 6 12 180 Ratnagiri 6 6 12 180 Satara 6 6 12 180 Kolhapur 6 6 12 180 Pune 6 6 12 180 40 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Annexes States Districts Number of NGP winners selected Households @ 15 per NGP winner Pre-2008 2008 Total Rajasthan Alwar 1 4 5 75 Churu 6 4 10 150 Jhunjhunu 5 4 9 135 Rajsamand 3 2 5 75 Sikar 3 4 7 105 Tripura South Tripura 7 6 13 195 West Tripura 7 8 15 225 Uttar Pradesh Pratapgarh 3 4 7 105 Kaushambi 3 6 9 135 Bareilly 6 2 8 120 LakhimpurKheri 5 6 11 165 Aligarh 4 6 10 150 Orraiya 5 6 11 165 Basti 7 6 13 195 Gorakhpur 7 7 14 210 Kannauj 3 6 9 135 Moradabad 7 7 14 210 Jaunpur 7 7 14 210 Bijnor 10 7 17 255 Mirzapur 7 7 14 210 Muzaffarnagar 9 6 15 225 Ghaziabad 7 7 14 210 West Bengal Bankura 6 6 12 180 Nadia 6 6 12 180 North 24 Paraganas 6 6 12 180 Medinipur 6 6 12 180 Total 332 332 664 9960 www.wsp.org 41 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Annexes Annex 2: Interview Guide Research Protocol Assessment of Processes that Drive Outcomes in TSC CONTENTS 1. Stakeholder Guide 2. Suggested Interview Structure 3. Component-wise Questionnaires COMPONENT 1: CATALYZING 1.1 Strategy for TSC Implementation 1.2 Institutional Structure and Capacity 1.3 Financing COMPONENT 2: IMPLEMENTING 2.1 Program Approach to Creating Demand 2.2 Supply Chain and Technology Promotion 2.3 Scaling Up COMPONENT 3: SUSTAINING 3.1 Subsidy 3.2 Monitoring 3.3 Rewards and Recognition 42 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Annexes 1. STAKEHOLDER GUIDE Stakeholder/Agency Whom to Interview District Water and Sanitation Mission • Collector/Chief Executive Officer • District Coordinator/Project Officer • District TSC Cell members (at least 1) • Optional - selected BDOs GP • Pradhan • GP members • Community members active in TSC • Households 2. SUGGESTED INTERVIEW STRUCTURE Introductions • Introductions • Appreciation for time • Purpose of interview • Confidential, won’t use name or other identifying information Components Use attached component-wise questionnaire guide. Closing • If you were to design the TSC and NGP, what is the one most important thing that you would recommend should be done? • Indication of when report will be ready. • Thanks for sparing time for this interview. 3. COMPONENT-WISE QUESTIONNAIRES COMPONENT 1: CATALYZING 1.1 STRATEGY FOR TSC IMPLEMENTATION 1. When was the TSC launched? Prior to the TSC, was there any rural sanitation program opera- tional? If yes, what was the experience and results achieved? 2. Is TSC a departure from the way conventional rural sanitation programs are implemented? If yes, how? 3. How are TSC Guidelines goals and approach interpreted? Is there a common understanding at different levels of implementation from state to Panchayat? 4. What priority is given to TSC vis-à-vis other development programs? What is the approach to TSC implementation? Does it follow the TSC Guidelines in letter and spirit? 5. What are the main challenges faced in implementing this strategy? Have there been any shifts in the district’s approach to implementing TSC? 6. What are the strengths and weaknesses of this strategy? What recommendations would you make to overcome any weaknesses? www.wsp.org 43 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Annexes 1.2 INSTITUTIONAL STRUCTURE AND CAPACITY 1. Describe the institutional structure to implement and monitor the TSC at the following levels: • District • Block • GP Ask: Who is responsible? What role/function do they play? What is their incentive to be involved with the TSC? Are named positions staffed? If not, why not, and since when? 2. Is there a dedicated TSC cell within the nodal agency or is this shared with other departments/ programs? How well is this cell functioning? In case the cell became redundant, what were the main reasons and how can these be overcome? 3. How is staff inducted into the TSC? Are they on deputation or recruited from the open market? What training is given to newly appointed staff? Does TSC progress have any bearing on staff performance review? Are there any incentives/disincentives to deliver on TSC goals? 4. Are any private contractors or NGOs hired to support implementation? What is their role? 5. Is there coordination of the nodal agency with other linked line departments/agencies? How many coordination mechanisms are still active and how well are these working? If any coordina- tion mechanism failed to work, what were the key factors and how can these be addressed? 6. In your opinion, are existing institutional arrangements adequate (in terms of staff, budget, capacity to implement, and so on) to meet the TSC goals within the program time frame? 1.3 FINANCING 1. What resources are available to implement the TSC in terms of: • Financial allocations16 Ask: Are fund flows predictable? Any bottlenecks or delays? How can these be overcome? • Non-financial resources, for example, training institutions, cross-subsidy for TSC activities from other rural development programs, staff on deputation, and so on. 2. What is the overall budget and breakdown of costs for implementing the TSC program? Does it cover ODF and other aspects of total sanitation, for example, Solid and Liquid Waste Management (SLWM)? Is this allocation adequate? 3. Is fund flow consistent or are there bottlenecks? How can any bottlenecks be overcome? 4. Are households mobilized to invest in their own toilet? How? COMPONENT 2: IMPLEMENTING 2.1 PROGRAM APPROACH TO CREATING DEMAND 1. What is the process of creating demand in a GP: • Who is the target audience at the community level? How was it identified? • What are the main messages used in demand creation? How were these identified? Does it focus on collective or individual behavior change? • What methods are used to create demand at the community level? How often? For example, Participatory Rural Appraisal (PRA) tools, house-to-house meetings, IEC tools (wall paint- ing, pamphlets, mass media, and so on), exposure visits, CLTS triggering, and so on. Collect samples/pictures if available. • How are reluctant/hard-to-reach households or areas tackled? 16 For various program components: start-up activities, administration, hardware and software related to household sanitation, and capital cost of non-human waste management and disposal. 44 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Annexes • Is the effectiveness of communication related activities tracked? If yes, how? 2. Who is responsible for creating demand at the community level? What support is provided to motivators towards travel and boarding/lodging for fieldwork? Are they given any training? Are there any incentives for motivators and how is their performance monitored? 3. In your opinion, what is the single most effective demand creation strategy or method and why? 4. What is the time frame for demand creation activities? 5. Are any communication activities undertaken in those areas that have become ODF or won the NGP? If yes, what type of activities are these and what are the messages? 2.2 SUPPLY CHAIN AND TECHNOLOGY PROMOTION 1. Is any particular technology model being promoted under the program? If yes, why and how? 2. Is there any RSM/Production Center(PC) functional? What is the role and experience? In case RSM/PC is not functioning, what are the main reasons and how can these be overcome? 3. Has there been any training program for masons or engineers under the TSC? If yes, when and how often? What was the objective and how many were trained? What were the results? 4. How would you rank the availability and consistent supply of the following sanitation products and services? [1 lowest to 5 highest] • Masonry services for toilet construction • Financing and transport of sanitary products • Availability of key sanitary products and spare-parts, for example, pans, pipes, slabs, and so on. Probe further based on ranking given 5. In your opinion, are existing sanitary products and services: • Affordable? • Of dependable quality? • Varieties are easily and consistently available? Ask: In case answer is negative to any of the above, ask why and what needs to be done to improve 2.3 SCALING UP 1. Is TSC implementation divided into phases? If yes, what indicators are used to select GPs in each phase? 2. Are there any targets for achievement of NGP? How were these set? 3. What is the strategy for reaching hard-to-reach areas or groups? Are demand creation efforts of similar quality in these cases? Are any efforts taken to facilitate the supply chain for sanitary products and materials in these areas? 4. Is the focus on total sanitation? Does it include households and institutions? Is there a plan to go from ODF to ODF+ activities? COMPONENT 3: SUSTAINING 3.1 SUBSIDY 1. When is subsidy released? What verification of usage, if any, is done before release? 2. What role does subsidy play in determining the choice of technology selected? 3. Are there any errors of inclusion or exclusion in the selection of households that receive the subsidy? Errors of inclusion mean that those households which are well able to afford the cost of toilet construction get included. Errors of exclusion means that those who really deserve the www.wsp.org 45 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation subsidy are not included. 4. Who is the subsidy amount given to? Private contractor, NGO, GP, individual household? 5. Is subsidy used as an incentive to reward a GP for achieving ODF status? 3.2 MONITORING 1. What is the system for monitoring the TSC at different levels: • State • District • Block • GP 2. What indicators are used? Who is responsible? What is the frequency of monitoring? 3. Do any incentives or disincentives exist to motivate interest in undertaking monitoring and using the results to drive performance? 4. To what extent is the (current or planned) process sufficient to monitor quality of processes, outcomes, identify gaps and weaknesses, and determine lessons learned and best practices? 5. Are results of the districts TSC MIS system being used to improve program performance? 6. Are NGP applications checked before forwarding for consideration? If yes, what process is followed? 3.3. REWARDS AND RECOGNITION 1. Are there any incentives or rewards for motivators or Pradhans to achieve ODF status? 2. Are there any rewards for block or district officials to prioritize TSC vis-à-vis other programs? 3. Is there any attempt to motivate GPs to sustain or go beyond ODF or NGP status? 4. Is there any competition-based prize for those GPs that sustain ODF or NGP? If yes, how does it work? What have been the results? 46 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation Abbreviations and Acronyms APL Above the Poverty Line BCC Behavior Change Communication BDO Block Development Officer BPL Below the Poverty Line CBO Community Based Organization CLTS Community-Led Total Sanitation CRSP Central Rural Sanitation Program CSO Country Status Overview DC District Collector or Deputy Commissioner DDWS Department of Drinking Water and Sanitation DP District Panchayat DSS District Sanitation Samiti EEA Enabling Environment Assessment GDP Gross Domestic Product GoI Government of India GP Gram Panchayat HRD Human Resource Development IEC Information, Education, and Communication M&E Monitoring and Evaluation MDWS Ministry of Drinking Water and Sanitation MIS Management Information System MoRD Ministry of Rural Development NGP Nirmal Gram Puraskar NGO Non-government Organization ODF Open Defecation Free O&M Operation and Maintenance PC Production Center PHED Public Health Engineering Department PPP Purchasing Power Parity PPS Probability Proportional to Size PRA Participatory Rural Appraisal PRI Panchayati Raj Institution RDD Rural Development Department RSM Rural Sanitary Mart SHG Self-Help Group SLWM Solid and Liquid Waste Management TSC Total Sanitation Campaign VEO Village Extension Officer Watsan Water and Sanitation WSP Water and Sanitation Program ZP Zila Panchayat/Parishad www.wsp.org 47 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation Glossary In order to ensure a common understanding of the concepts District Total Sanitation Campaign (TSC) Cell: Under and terms used in the report, some of which are specific to the GoI’s national rural sanitation program, a dedicated the Indian rural sanitation context, the following defini- cell can be set up at the district level to support manage- tions are provided: ment of different components of the program. This cell typically comprises specialists in human resources, moni- [Above/Below] Poverty Line: To measure poverty, it is toring, training, and communications who are either hired usual to look at the level of personal expenditure or income as consultants or deputed from different departments in the required to satisfy a minimum consumption level, asso- district administration. ciated with a commodity bundle that yields a minimum calorie intake per day. The Planning Commission of the Information, Education, Communication: IEC is the term Government of India (GoI) uses a food adequacy norm of used to describe software activities that support communi- 2,400 to 2,100 kilo calories per capita per day, to define cation and capacity development around the desired behav- state-specific poverty lines for rural and urban areas respec- iors promoted by the national rural sanitation program, tively. Energy intake norms are then transformed into the such as using a toilet, washing hands with soap, safe disposal price of a food basket to realize this, which is then trans- of garbage and waste water, and so on. IEC related activities lated into poverty lines in terms of food expenditure. These include mass media communication as well as inter-person- poverty lines are then applied on India’s National Sample al communication/outreach. Survey Organisation’s household consumer expenditure distribution to estimate the proportion and number of poor Nirmal Gram Puraskar (Clean Village Prize): NGP is an at the state level. incentive program introduced by GoI that gives a cash prize to any local government that achieves community- Chief Executive Officer of a District: The CEO of a wide total sanitation. More than a fiscal incentive, the district is an official responsible for coordinating adminis- award carries tremendous prestige as it is presented by the tration of all departments excluding revenue and law and President of India to block- and district-level winners and order in a district in the Indian government’s administra- by high ranking state dignitaries to village-level winners. tive structure. Nirmal Bharat Abhiyan: Literally, Clean India Campaign. Anganwadi: Pre-school or crèche, an initiative promoted Nirmal Bharat Abhiyan is the name given to the restruc- under the Integrated Child Development Scheme (ICDS) tured TSC, the second national rural sanitation program of the Government of India. of the Government of India which comes to a close in 2012. As the TSC objectives of universal rural sanitation Block Development Officer: A Block Development Officer coverage have not yet been achieved, this program has (BDO) is an official in charge of a block or sub-district in been restructured as the Nirmal Bharat Abhiyan, India’s the Indian government’s administrative structure. A BDO third national rural sanitation program, which is slated to coordinates and monitors planning and implementation of end in 2022. all government programs at the block level. Panchayati Raj Institution: The term ‘Panchayat’ liter- District Collector (or District Magistrate or Deputy ally means ‘council of five [wise and respected leaders]’ Commissioner): Chief administrative and revenue offi- and ‘Raj’ means governance. Traditionally, these coun- cer of a district in the Indian government’s administrative cils settled disputes between individuals and villages. The structure. modern Indian government has adopted this traditional 48 Global Scaling Up Sanitation Project WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation term as a name for its initiative to decentralize certain Total Sanitation: A community-wide approach based on administrative functions to elected local bodies at village, participatory principles, which seeks to achieve not only block, and district levels. It is called a Gram Panchayatat 100 percent open defecation free (ODF) communities, that the village level, Panchayat Samiti at the block level, and is,all households have access to and use a functional toilet at Zila Parishadat the district level. all times, but also broader environmental sanitation objec- tives such as promotion of improved hygiene behaviors Pradhan: The elected head of the local self-government unit such as hand washing with soap and safe disposal of non- at the village cluster level. In some states, a Pradhan is also human solid and liquid waste. Under the NGP incentive known as a Mukhiya or a Sarpanch (also see: Panchayati program of the GoI, a village is considered ODF only when Raj Institution). safe disposal of human fecal matter is ensured at all times. This means that even if every household and institution Rural Sanitary Mart (RSM): Under GoI’s TSC, RSMs are (school, pre-school, and so on) in a village has a toilet, the envisaged as retail outlets for construction material and village would not be considered ODF unless: other articles required to build sanitary facilities, primarily • Household, institutional, and public toilets are func- toilets. They are expected to provide information on locally tional and found to be in use on a continuous basis; available options for toilets, direct households to trained and masons, and promote improved hygiene practices. • Infant feces are disposed safely. www.wsp.org 49 WSP Linking Service Delivery Processes and Outcomes in Rural Sanitation | Executive Summary www.wsp.org 51 Water and Sanitation Program 55 Lodi Estate, New Delhi 110 003, India Phone: (91-11) 24690488, 24690489 Fax: (91-11) 24628250 E-mail: wspsa@worldbank.org Web site: www.wsp.org