E-QLT

As per the recent IPCC report, half of the world’s population lives in Climate ‘Danger Zones’ with risks to health and life making them vulnerable. In India according to an Oxfam India 2023 report, India still has the highest number of poor at 228.9 million, making them highly vulnerable. While the top 5% own nearly 62% and the top 1% own nearly 40.6%, the majority of the country is left behind. Since The Covid-19 pandemic, the bottom 50% have seen a decline in their income share to an estimated 13% and have less than 3% of the total wealth in 2020. Reports finds that inequities in health, education, household characteristics and the labour market increases the vulnerability of populations, driving them towards multidimensional poverty. 

Vulnerability is a wicked problem which makes it difficult to comprehend and even more difficult to design interventions to reduce it. Stakeholders like insurance companies, governments departments at various level, community-based organisations, community groups play an important role in reduction of vulnerability. However, despite plethora of social protection interventions, the reduction of vulnerability has not kept in pace with the growing vulnerable populations. The reason for the same are: 

1. Stakeholders lack approaches that allow them to understand who the vulnerable people are and when they become vulnerable and measure the differences in vulnerability at different scales. 

2. They also lack a way through where they can test which interventions could work in reduction of vulnerability. 

3. At present there is no way for these stakeholders to understand how their various interventions can come together and cumulatively help in reduction of vulnerability. This leads to increase in cost for the programs like insurance, as it overestimates vulnerability in cases and makes it unaffordable. 

We seek to address these gaps and bring together the diverse stakeholders in our social security simulation platform E-QLT. Through the platform these stakeholders can see how vulnerable their target audiences are, test interventions, simulate shocks, and plan for resilience. They can use existing datasets like census, socio-economic survey, national health survey, scheme databases, etc. or add their own datasets. Through the platform they can also understand the combined impact of different interventions and look at bringing together different interventions or test new interventions.

E-QLT is a simulation platform that enables stakeholders like insurance companies, government, community-based organisation to leverage data to better understand vulnerability of households and test interventions to help reduce vulnerability. 

The simulation platform models how a household makes choices with respect to their income and expenditure and their coping strategies during stresses like variable income, shocks, loss, etc. The choices the household makes as coping strategies take different forms, E.g., during income shocks, 

1. They could use their savings, which reduces their potential to deal with more shocks. 

2. They could do distress selling of assets, which reduces their long-term financial assets. 

3. They could resort to debt, leading to increased debt burden and in some cases lead to debt trap. 

4. Or they can cut back on different expenditures like nutrition, education, etc., and this could lead to impact on health or educational outcomes in the long term. 

The platform allows users to see how household's choices during the shock leads to impacts on different aspects of their lives and makes them more vulnerable. To understand the vulnerability of the household, we convert the results of the model to a simple to use indicator, Social Protection Score (SPS). SPS is a score between 0-300; below 300 the households are vulnerable, and 0 refers to critical vulnerability. The 300 score is further subdivided into indicators which represent health, education, and finance of the household, thus allowing for understanding where the vulnerability stems from. The platform further helps test various interventions and quantifies the interventions through increase in SPS, and highlights the gaps left. Through this, they can look at how different interventions, both existing and new, can be brought together to promote resilience.

E-QLT platform is aimed to enable key stakeholders like private sector actors (insurance and credit companies), governments, community-based organisation and community groups towards designing interventions that lead to reduction in vulnerability of groups that they work with. These interventions could be insurance, new models of finance, cash transfers, in-kind transfers, etc. 

Through this we seek to reach the most vulnerable population in the Global South and work towards building resilience against poverty, climate disasters, health shocks, etc. The platform also seeks to bring diverse actors together and accelerate the progress towards resilience. 

At present, despite plethora of social protection interventions, the progress towards reduction of vulnerability has been limited. This leads to millions of people without any social protection net or with inadequate social protection. By measuring vulnerability and the contribution of various social protection measures, we seek to highlight the gap and promote pathways that help bridge this gap.

The team members who are building and scaling E-QLT come from diverse backgrounds like social science, urban planning, computer science, economics, journalism, etc. The tool leverages with the team members' decade long experience of working in development sector in the domains of urban poverty, disaster management, housing, etc. This enables the technology to be grounded in research and evidence. 

The organisation follows participative approach towards tool building, and leverages partnerships to ensure the tool building is relevant to the context. E-QLT is built in close partnership with community-based organisations to understand the community's needs. 

The Team Lead, Yashwin Iddya belongs to one of the backward castes in India and recognizes the role various government interventions like public schools, affirmative action in higher education, etc. played in enabling growth in their lives. This is their motivation to ensure design of tools that could enable stakeholders like government and private sector, to design relevant products and schemes enabling expansion of the social protection net for all.

Through different pilots, we have worked with Community based Organisations, Research Organisations, and Development Organisations. The E-QLT platform enables these organisations to measure vulnerability of the communities they serve better, and to test the contribution of various social protection schemes more effectively. We have tested the platform in five geographies: 

1. Kerala: Here using secondary household level data of 400+ households and a typology-based approach, we demonstrated how various cash transfer schemes help in reduction of vulnerability of household and what gaps still exist towards ensuring no vulnerability. This was used as a research study in collaboration with Dvara Research Organisation. 

2. Ambujwadi, Maharashtra: Using primary data of 86 migrants in a locality in Mumbai, various government schemes were tested for their contribution towards reduction of vulnerability of these households. This was used both for advocacy and for program planning by the community based organisation Youth for Unity for Voluntary Action. 

3. Panvel, Kagal and Chikhaldara, Maharashtra: These places are representative of Urban, Rural and Tribal areas, where through a sample of 155 households, vulnerability assessment of these households was conducted, and contribution of various schemes was measured. It was part of advocacy in collaboration with Youth for Unity and Voluntary Action and UNICEF India.

We see potential for the solution we built to be scaled globally and used in different geographies and institutional contexts. We seek to build the business model that considers different user groups, e.g., how do you make the platform for community, ensuring affordability, while still making the platform financially sustainable. 

With SOLVE's experience over the years through growing different solutions, we seek to learn more and develop a business model that enables sustainability and resilience of the solution. Given the solution is primarily a decision support tool, we want guidance on how to position it in different markets and see how the features of the model could be updated or new ways to look at how the platform could be conceptualized.

Understanding how households become vulnerable, and how various social protection measures help in reducing vulnerability is crucial to designing better interventions like insurance, schemes, etc. Based on review of literature, following gaps are identified: 

1. Approaches like Multidimensional Poverty Index (MDPI), Poverty line, etc. don't account for temporal nature of vulnerability. 

2. Approaches like MDPI lack interconnection among variables like health, education and finance. 

3. The approaches lack a mechanism to integrate the role of shocks and nature of shocks, and how that increases vulnerability, and, how the contribution of various schemes/interventions are in reduction of vulnerability 

4. Most indicators look at a macro level and fail to account for differences between households. 

To address the above issues, we built upon existing approach of MDPI, and used a simulation approach that models the household, and measures their vulnerability. The simulation approach at household level, allows for looking at differences at household level, while allowing for aggregation at a regional level to address the issue of scale. Once the vulnerability is understood, one can test different existing schemes/intervention in isolation or together, allowing to look at the compound effect of the social protection net. 

The most innovative feature of the tool is that it allows the testing of different scenarios like disasters (natural and man-made), health shocks, etc. and test different interventions. These interventions can range from cash transfers, in-kind transfers, to innovative new financial mechanism and insurance. The platform is envisioned as a space which could foster collaboration between government, private sector, community-based organisations towards a holistic approach of vulnerability reduction.

We are currently in phase 1 of our growth journey, where we seek to have pilots with diverse use cases and build a platform integrating the learnings. We have had successful pilots with community based organisations, research organisations and development organisations. In the next year, we plan to pilot E-QLT with insurance companies, policymakers and community groups. With this we seek to build a platform that enables different stakeholder to come together and work towards reducing vulnerability in a holistic manner through new products/interventions. We seek to do this through partnerships that will build the base for future collaboration. 

In the next 5 years, we seek to ensure that the platform can scale to different countries in the global south; this will be through partnerships with organisations who are embedded in the region in different countries.

To measure our progress, we are looking at 2 sets of indicators: 

First level Indicators: We are focussing on several different stakeholders who are using the platform across different groups like private sector, government, community-based organisation, community groups. These are the different key stakeholders whose efforts play an important role in reduction of vulnerability. 

Second level indicators: Here the focus is on the number of people each of the stakeholders could impact, the types of marginalised groups, the types of interventions being planned and tested, and potential scope of reduction of vulnerability of population in these geographies. We seek to measure the progress of these indicators with region specific baselines like poverty headcount, vulnerable population count, etc.

The “Theory of Change” pathway asks for one to establish linear connections between interventions and their outcomes. But problem solving in complex social contexts rarely works that linearly. For example, it is nearly impossible to say exactly what the root causes of poverty might be, the root causes might vary greatly from cultures to communities, and even between households within the same communities. Similarly, interventions to lift people out of poverty might have vastly different outcomes, from household to household, depending on the composition of the household, what other interventions they have access to, the unique mindsets of people within the households. EQLT as a tool shows us exactly this in a scientific, evidence-based way – that the same welfare scheme built has different impacts on different households, and how does one use this insight to improve social protection better. 

So, we won’t be articulating our Theory of Change here. Instead, we prefer to view social problems from the lens of “wicked problems,” that have many interdependent causes, existing within complex systems where interventions have diverse outcomes. These factors are also constantly in flux. Therefore, solving wicked problems requires a deep understanding of the stakeholders involved, and a systems thinking approach. E-QLT is based on research of different approaches towards poverty reduction, stemming from economics, development studies, social science, and anthropology along with approaches towards understanding vulnerability. E-QLT models the household using 1000+ variables to understand the social vulnerability of a household and lets stakeholders test various interventions to lift them out of poverty and reduce their vulnerability. Taking these many variables into account, instead of a straightforward intervention and trying to predict its outcome, is what makes EQLT the first tech tool of its kind that measures poverty from a systems lens.

The E-QLT simulation platform consists of a model and web-based architecture for data collection and visualisation. 

The core of E-QLT model uses a system dynamics model that models a household using 1000+ variables, that is integrated with web-based services for data gathering, cleaning, storage and visualisation. The result of the simulation is converted to a Social Protection Score, to allow for easy comprehension of vulnerability. The model also consists of using statistical methods to aggregate and present data at different granularity. 

The web platform provides end users the ability to link existing databases like census data, socio-economic survey, GIS data, disaster datasets, scheme-related data etc. and also to collect primary data, if needed. This allows it to both leverage existing datasets, or to collect new data. The visualisation uses different libraries to allow for easy understanding of the results to promote actions.

E-QLT model is a simulation based on Sustainable Livelihoods Approach. The result from the simulation is converted to a Social Protection Score which builds upon standard indices like Poverty line, MDPI, Incidence of Indebtedness, Life expectancy, etc. During our pilot we compared results from our simulation platform, the Social Protection Score, with indicators like International Poverty Line, Tendulkar Poverty line (specific to India) and sub-indicators under SDG 1, 2 and 4. This allowed us to compare our approach with standard approaches and demonstrate how we help address gaps. The report for the same is attached here:
https://sitkgok.com/public/fil...

Given the complex nature of the problem, we have an inter-disciplinary team coming from diverse backgrounds. The team consists of members from studies of urban planning, economics, social studies, arts, computer science, among others. They also come from diverse socio-economic backgrounds. As an organisation we seek to promote people from diverse economic backgrounds through ensuring at the entry level, the pay we offer is above the current market rates and is also adequate to meet most of the needs living in the city. We also promote a hybrid working culture to enable greater geographic diversity of people who work with us. 

Given the diverse context of vulnerability, for each project we work with partners who are embedded in the context. For this solution during its development in different stages, we have partnered with Dvara Research Foundation (research institute), Youth for Unity and Voluntary Action (community-based organisation) and UNICEF India. We are also in in process of working with policy makers.

We anticipate different revenue models in different phases of the growth of the platform. We also plan to cross subsidize community groups and community-based organisations through revenue from grants and user fees from other stakeholders. We see the following approaches in different phases of the growth. We are presently mid-way in Phase 1. 

Phase 1: Platform Development and Scaling 

Phase 1 is conceptualised as involving platform development, pilots with different use cases, and scaling the platform to diverse contexts. In this phase we seek to use grants and service contracts approach, to develop and test different aspects of the technology with different users, and to incrementally build features, and add modularity in the technology. At the end of phase 1, we hope to achieve maturity with technology in terms of understanding how we add value to different users, and how do we position the product. We seek to also have evidence of the solution for different user groups. 

Phase 2: Platform Maturity and Embedding in the ecosystem 

Phase 2 is focussed on growing the use of the platform among different user groups, and seeks to embed the platform with the users, to ensure sustainability. We anticipate two types of use of the platform at this stage. One, where the users come to the platform that is run and maintained by us; for this we seek to charge usage-based fees to cover the cost of running the platform. In the second, we anticipate certain users would want a customised version of the platform. E.g., one of the target groups are Policy makers; different government departments would want to embed the use of the tool within them, while they plan new schemes to improve social protection and resilience in communities. For such users we plan to charge a deployment fee, and the regular maintenance and upkeep of the version would be done by the stakeholder. Our role would then be to modify the platform when the need arises. The cost of deployment could either be financed by the stakeholder or through grants.

We see the following approaches in different phases of the growth. We are presently mid-way in Phase 1 

Phase 1: Platform Development and Scaling 

Phase 1 is conceptualised as involving platform development, pilots with different use cases, and scaling platform to diverse contexts. In this phase we seek to use grants and service contracts approach, to develop and test different aspects of the technology with different users, and to incrementally build features, and add modularity in the technology. At the end of phase 1, we hope to achieve maturity with technology in terms of understanding how we add value to different users, and how do we position the product. We seek to also have evidence of the solution for different user groups. 

Phase 2: Platform Maturity and Embedding in the ecosystem 

Phase 2 is focussed on growing the use of the platform among different user groups, and seek to embed the platform with the users, to ensure sustainability. We anticipate two types of usage of the platform at this stage. One, where the users come to the platform that is run and maintained by us; for this we seek to charge usage-based fees to cover the cost of running the platform. In the second, we anticipate certain users would want a customised version of the platform. E.g., one of the target groups are Policy makers; different government department would want to embed the use of the tool within them, while they plan new schemes to improve social protection and resilience in communities. For such users we plan to charge a deployment fee, and the regular maintenance and upkeep of the version would be done by the stakeholder. Our role would then be to modify the platform when the need arises. The cost of deployment could either be financed by the stakeholder or through grants. 

We understand out approach could have trade-off in terms of speed of uptake, and we are open to exploring other models.

As described above, we are presently on phase 1 of the journey, where we are using grants and service contracts to fund platform development, test pilots and incrementally grow both the technology and user base. In our journey an initial grant by The Nudge Foundation, helped us create a prototype of the vulnerability model, and then through two service contracts and a further grant we were able to have pilots with Dvara Research Foundation (research organisation), Youth for Unity and Voluntary Action (community-based organisation) and UNICEF India (development organisation). The technology also underwent increase in maturity, where the vulnerability model was further refined, and end to end automation was added to allow for streamlining the use of the tool at a larger scale.