Bandhu

As the world urbanizes rapidly, driven by failing rural economies whose collapse is exacerbated by climate change, the process of choosing to migrate remains full of blind turns. No single entity--employers, middlemen, landlords, and workers--has perspective on the entire chain. One wrong step for a poor family can have serious lifelong consequences.

Our solution, Bandhu, brings all these entities together on a single mobile phone-based platform. It minimizes these information asymmetries so that low-income migrants can make migration decisions with assurance. Bandhu creates three-way matches between employers looking for low/moderate-skill workers, potential migrants/workers, and owners/occupants offering informal affordable housing on rent.

Bandhu could smoothen the urbanization process for billions of people across the developing world. Bandhu can make urbanization more inclusive and accessible to the rural poor, refugees, and women fettered by social and economic constraints, who struggle to improve their lives through greater social and economic independence.

Worldwide, millions of people are migrating in search of better jobs and housing. These migrations are exacerbated by factors including climate change-induced crop failure and subsequent lack of rural economic opportunities. Such patterns are particularly acute in India, where 70% of 1.4 billion people live in rural areas. Rural workers often rush to cities out of economic desperation. Consequently, they are neither full-time residents of a city nor of a village, but constantly in transition. Urban centers fail to provide inclusive environments for such a dynamic group, and workers often find themselves in exploitative arrangements. The only connection between rural and urban economies are the middlemen, who capitalize on their rural social ties and urban economic connections--often capturing 25-40% of the value in the migration process. There exist extensive information asymmetries between rural and urban networks. In this environment, 150 million Indians undergo this distress-driven, cyclical migration process every year; one wrong step for a migrant can have serious lifelong consequences.

We primarily serve low-wage migrants/prospective migrants between the ages of 18 to 40. Many take up unskilled or semi-skilled construction labor work. They are prone to exploitation by employers and middlemen, and often the untenable costs of urban living force them to go back to their villages. 

We have performed over 160 interviews and feature team members experienced in India’s construction industry. For the past two months, we have been collaborating with a local NGO (Aasmaan) that has worked in particular informal worker settlements in Ahmedabad, India for the past ten years. We are running participatory pre-pilots with these communities; they respond to mock-ups of our app, ask questions, voice concerns, and provide feedback. We are incorporating their input and onboarding workers, whose profiles will be pre-loaded into a market-ready prototype by September 2019.

This solution establishes a system of trust for these workers, enabling them to make rational livelihood decisions. We address the factors perpetuating exploitation through a two-pronged approach: 1) by improving information availability, giving this underserved population the tools to become more economically resilient in the face of constant change; and 2) by altering the incentives for the middleman to produce a more inclusive system.

Our mobile phone platform, Bandhu, maximizes transparency in the migration process so that low-income migrants can make life decisions with assurance. It does so by creating matches between employers looking for low/moderate-skill workers, potential migrants/workers, and owners/occupants offering informal affordable housing on rent. By providing potential migrants with competing offers of bundled employment and housing—for similar net wages—Bandhu minimizes risks for the worker while still leaving the final choice up to him/her, based on personal preferences like take-home pay, distance traveled from home to work, and other factors.

Bandhu takes advantage of India’s high rates of mobile phone coverage, one of the lowest worldwide data rates, and rapidly-increasing smartphone penetration. 300 million Indians have smartphones, and 70% of rural households have mobile phone access. Unlike the West, where a mobile phone is a personal asset, mobile phones in rural India are commonly considered family or community assets, increasing our potential to reach more users.

This three-way match system will require a matching algorithm that accounts for individual worker needs. This algorithm will suggest possible job-housing combinations given a worker’s skills, schedule, and personal preferences. These preferences include a worker’s expected wages, willingness to travel, proximity of housing to public schools for workers with families, neighborhood safety (using proxies like level of street-lighting visible through satellite imagery), and food affordability (i.e. presence of street vendors). For example, it would allow a migrating family to choose from a set of neighborhood options that are safe for children, having high levels of "eyes on the street" and within walking distance to public schools. 

As more users join our app, we will have a larger pool of data to analyze using machine-learning and recommendation algorithms, allowing us to make better suggestions based on the preferences, behavior, and psychological needs of workers from different backgrounds (for example, studies have shown that migrant workers who have reduced opportunities to speak in their mother tongue can have adverse impacts on their mental health).

Our platform will also integrate widely-used mobile payment systems and user authentication methods held by the government and by NGOs that we intend to partner with. In order to boost incomes, we plan to provide skills training for workers who consistently receive low ratings from employers, through our relations with these partner NGOs.

While global narratives postulate utopian smart cities, they ignore the need for “just” transitions from rural distress to these imagined utopias. We focus on designing just and inclusive transitions through the following:  

1. We see low-wage seasonal migrants as having dynamic, rather than fixed, characteristics. This contrasts with the approach of most public- and private-sector entities, which attribute fixed characteristics to these populations and disregard the highly dynamic and transitory nature of these workers’ lives. Our approach creates monetary value from such frequent and sustained movements (transactions), and incidentally collects data to enable further insights into migrant needs. This previously unrecorded data is essential for making improved local and large-scale public policy decisions related to infrastructure development (i.e., transportation and temporary housing). 

2. Unlike aggregator platforms that emphasize “disruption” as innovation, we believe in “inclusion” of existing actors and reorienting their incentives. We leverage existing social capital into financial capital that can be more equitably redistributed. This is done through two pivotal actors: middlemen, and local street vendors. 

   1. Middlemen enjoy the trust of workers, and their network translates to powerful social capital. While maintaining these pivotal connectors between rural and urban spheres, we reward them commensurate to their capacity-building for each worker they refer onto the platform. 

   2. Local tea and street vendors occupy a trusted position on the ground, typically also providing lines of credit for local workers and knowing which workers are most reliable. Thus, these partners are invaluable to us as a bridge to onboarding in their communities.

Our technology helps transform social capital into financial capital that can be redistributed among underserved communities. 

Bandhu relies on a novel business model that converts community knowledge into digital insights and recommendations. It converts the street knowledge and networks of the middlemen into a monetizable long-term revenue resource for them, wherein we nudge and realign rather than disrupt. We begin by working with local partners to run surveys of various low-wage groups in the construction, manufacturing, and delivery sectors. 

- We build a database of these answers to gain knowledge about worker skills and schedules, typical migration patterns, job priorities/concerns, and housing preferences

- We will supplement these stated preferences with live data that captures revealed preferences. 

- These insights will be used to build a robust machine-learning algorithm that is responsive in real-time and makes optimized recommendations for job-housing bundles for each worker. 

We recognize that survey sample size and algorithmic bias are possible risks, and plan to cross-check these by working with our on-ground partners to survey additional workers and leverage these partners’ existing qualitative knowledge of labor and migrant networks. In doing so, we will have created a platform that helps transform social capital into insights that create financial capital for these workers. 

Activity: Adequate information from trusted sources, enabling workers to compare across various employment and housing bundles in cities within an overnight travel distance from their native villages. 

(Evidence: Regression results from Rushil’s master’s thesis)

Short-Term Outcomes (18 months)

Evidence: Field surveys/analyses

• Workers find better and stable jobs, secure rental housing, and immediate improvements in family income

• Workers have stability of being in one place; they can travel to and from native villages at short notice

• Workers can make individual migration decisions with assurance

• Reduced inefficiencies for employers/landlords: more reliable workers/tenants

• Reduced search costs for middlemen

Medium-Term Outcomes (36-60 months) 

Evidence: Data collection/analyses on limited capital and consequent tradeoffs

• Workers choose to move to cities offering a better wage differential and minimal loss of social capital

• Workers can pay for their children to complete schooling, rather than a disrupted education altering between English and vernacular mediums between cities and villages

• Workers have better access to healthcare in urban centers

• Decline in predatory behavior of middlemen

Long-term outcomes (10 years): 

Evidence: Interviews, academic research

• Exponential inter-generational upward mobility: formal sector jobs provide greater financial inclusion and access to housing mortgages

• Security of housing tenure/asset building through ownership, enabled through access to home loans/mortgages

• More inclusive cities from a migrant’s perspective; research showed education/financial inclusion leads to stable housing and greater urban socioeconomic inclusion

• Less pressure and infrastructure/waste management burden on cities which are already at capacity, leading to more equitable and healthy regional agglomerations

We have onboarded 70 workers to our platform, increasing at the rate of 20 workers per week, and anticipate onboarding 40 employers by September 2019. One year from now, our system of referrals and our network will have brought 15,000 people onto our platform, including 720 employers. In five years, we anticipate a user base of 2.5 million—over 50% of the 4 million potential customers in the Ahmedabad-Mumbai corridor, our area of focus for the MVP. We are partnering with an NGO that serves approximately 1,000 individuals from low-income communities, representing a strong network to initially draw from.

Our solution is sustainable and inclusive in its approach to preserving but improving the existing structure of the informal labor market.

We are starting with a pilot based on Ahmedabad, focusing on building only the employer side of our platform for construction workers who are hired from labor nakas (daily wage markets). Their cost of trying our app is very low compared to their current activity of waiting at the labor squares every day. September to December 2019 is dedicated to developing and deploying our platform, and proving that it can successfully connect workers and employers (directly and through middlemen) in this initial phase. By December, we will have a daily database of worker-employer connections, the quality of those connections, and stakeholder concerns. Immediately thereafter, we will begin building the database for the housing side, and begin expanding out from labor nakas in Ahmedabad to other areas of the city.

We also intend to scale our services through partnerships with local NGOs, in Gujarat and elsewhere. We have already identified NGOs that help train migrant workers and give them unique ID cards that validate their identities and roles, and expect to partner with these NGOs to scale up our business and securely/ethically store worker data.

In the next five years, we intend to further develop our relationships with these groups, make the housing side more robust, attract more investors, and expand to other regions of India, targeting 10 million potential users.

Financial barriers: Many investors are awaiting a tested prototype until they decide to invest. Over the next year, we will be testing the prototype with an ever-increasing pool of stakeholders. Throughout this process, we will need to navigate both investor expectations and developer capabilities. The success of our pitches to investors will increasingly rely on the platform’s real-time performance. 

Legal barriers: As we scale, we must ensure quality and safety so users are held accountable and using our platform as intended. Informal markets also do attract illicit activity, and we will need to ensure these activities do not enter our platform. 

Cultural/linguistic barriers: Each labor market in each city is slightly different, and each state in India has its own primary languages. There is a different word for each profession and what is expected within that profession, making the process of building a user base in each area somewhat complex and customized. 

Market barriers: Currently no other business is in this market space, but we anticipate competition as soon as we release our product. We can file for IP only once our algorithm is ready following the analysis of substantial data points.

Financial barrier mitigation:  We will prioritize the development side of the platform to ensure the app’s quality regarding the interface and back-end matching/recommendation algorithms. We are actively pitching and pursuing funding sources and angel investors.

Legal barrier mitigation: We will work with local partners to source reliable and trustworthy middlemen and employers at the start, then build our user base in a controlled way to quality check. We will supplement the digital ratings with on-the-ground feedback to understand the full picture of our platform’s effects. For example, to safeguard against scenarios such as an employer compelling a worker to provide a high rating despite the worker’s unhappiness with the job, we will build a strong support/help team which allows individuals to anonymously report bad behavior on the platform without fear of repercussions.

Cultural/linguistic barrier mitigation: We are already rolling out the app in different languages (English, Hindi, Gujarati), and will work with well-established local partners who have indigenous knowledge to ensure a seamless and localized user experience. 

Market barrier mitigation: Loyalty and rewards strategies will keep users engaged with our app and out-scale the competition, creating enough trust that users will resist switching to a competitor. We will keep our marketing low-profile, and file for IP before we focus on conventional marketing. We will also use our marketing and branding to introduce Bandhu into the household vernacular.

We have two full-time founders, two part-time founders, and field operations including one full-time coordinator and five full-time interns/employees based in India.

Our co-founders include one on-ground partner and three Urban Planning Master’s graduates who each completed India-based fieldwork for their theses, interviewing 160 workers and employers in various informal-sector capacities. 

Rushil brings over four years of experience working on construction sites alongside migrant workers in India (Bandhu’s potential customers). He also carried out extensive fieldwork and market research for Bandhu on a Seed Grant from the Legatum Center at MIT. In other roles, Rushil has consulted for governments on such topics as slum redevelopment, affordable housing policy, and completed pro-bono work for postwar development initiatives for refugees in Sri Lanka. As a Fulbright Fellow at MIT, he focused on real estate, housing, and public infrastructure, including client based projects in Peru and Mexico.

Jacob brings a high-level strategic approach from five years of experience as a management consultant, specializing in big data analytics and value chain analysis. He speaks conversational Hindi and has traveled to India several times in various roles, including volunteering on construction sites with low-caste communities. At MIT, he studied opportunities for economic resiliency in rural Gujarat and received travel grants to study solutions for Mumbai’s informal waste management sector. His interests focus on urban information systems and applying technology solutions to real-world problems. 

Karthik, also an MIT alum, works in data science and analytics consulting and has a strong urban transportation behavior focus.

Viral is leading field operations in Ahmedabad, India, and brings over ten years’ experience of working with these communities through the nonprofit he co-founded.

Bandhu is currently partnering with Aasmaan NGO in Ahmedabad, India. Aasmaan has been serving over 1000 low-income individuals in the slum neighborhoods of Ahmedabad for the past ten years. As a key implementation partner, they are spearheading our pre-pilot in Ahmedabad, gathering user feedback on our idea using field surveys and in-person app testing. They are also onboarding the workers in the communities they work in.

Our business model creates a three-way match between workers, employers, and landlords on the same platform while also incorporating the middlemen.

Workers get access to the most optimal employment/housing bundle and do not have to pay anything for these services. They get access to secure housing tenure and relief from predatory commissions/brokerages.

Employers in the medium- and small-enterprise sector get access to skilled workers optimal for their requirements. They are assured better-quality workers with lower attrition rates, while still having flexibility to suit their variable project demands. In our platform, they pay a 10% service fee on a completely transparent system, as compared to the 25% they pay currently.

Landlords get access to a wider market, pay much-reduced brokerage fees, and have assured tenancy matching several social criteria.

Middlemen, who currently must take high risks and hustle to provide workers to employers, find an easier way through Bandhu to get assured commissions with reduced risk. While our platform reduces their commission per individual transaction, net income for the most trusted middlemen increases as they get access to a much larger market and collect regular income through a revenue-sharing model. By tying their incentives to worker quality and retention/delinquency rates, Bandhu innovatively converts the middlemen’s social capital and networks into a more equitable system that also earns them a more stable income. In the long term, this model will buy them time to diversify from the brokerage business, which is currently at high risk of disruption with India’s expanding tech ecosystem.

We would fall under a market intermediary model. The venture would be kept financially sustainable for the first three years with investment capital and grants. We have already received $31,500 in  prizes and grants at MIT. We are now looking to raise investment capital and eligible grants for social impact tech innovation. We expect to have positive operating profits by year four.

The service fee/commission received through the platform, as well as revenues from float and micro-lending enabled by the centralized payment processing, would help offset operating losses in the first three years, and becomes a key revenue source in the long run as the platform scales.

1. Visibility - As a globally-recognized program, MIT Solve would provide us with enhanced visibility to larger international networks. So far, we have gained much from the U.S.-based startup ecosystem, but have had limited opportunities to connect with larger networks outside the U.S. The visibility Solve gives us will help us attract top talent and build on local partnerships. 

2. Partnerships - MIT Solve stands apart in attracting mission-driven investors and partners. As much of our success will be contingent on robust local and international partnerships, Solve’s access to this strong global network will be invaluable to our mission. 

3. Legitimacy - MIT Solve would connect us to prominent institutions and individuals, lending our venture legitimacy in places where reputation and association are prerequisites for success. 

4. Mentoring - MIT Solve’s 12 months of close mentoring would significantly build on our existing successes, as we develop and market our ideas to a much wider audience. Solve is well-known for its focus on social impact, and we particularly hope to receive mentorship in taking our venture beyond the prototype stage. Furthermore, we would benefit from guidance from Solve’s mentors on how to scale up responsibly and with clear ethical milestones. 

5. Investors - Investors associated with MIT Solve already have a clear social-impact focus, and are often tied to much larger institutional networks. As part of our immediate goals is fundraising, these connections would be invaluable in scaling our enterprise.

We would like to partner with international non-profit organizations such as the Bill and Melinda Gates Foundation, which is a large mission-driven organization with a strong presence and reputation in India. Their connections within India would be invaluable as we begin expanding our platform’s user base and gaining the trust of local organizations.

Such local organizations, like Aajeevika in Ahmedabad, are focused on labor rights and upliftment for workers and have the trust of their communities. They would be useful partners in scaling responsibly, and could also provide guidance on local cultural nuances. We would see them as customer acquisition and skill-building partners.

Multilaterals such as IFC, IOM, and others interested in addressing climate change, refugees, and the promotion of SDGs would also be critical partners to enhancing the legitimacy of our venture and providing access to grants and affordable capital.

The Omidyar Network has been a trailblazer in the field of responsible impact investing.  Receiving support from such an organization would go far beyond a simple monetary impact. It would increase our credibility and expand our possibilities for rapid growth not just in India, but in other developing countries.

Bandhu is a cross-disciplinary technology solution to current inefficiencies and unjust practices plaguing labor migration markets worldwide. It helps transform social capital into financial capital that can be redistributed among underserved communities. Our approach is to nudge and realign rather than disrupt, which preserves the networks of local actors while redirecting their activities to more equitable ends. 

Our model combines qualitative insights collected on-ground with machine-learning algorithms to suggest ideal job-housing bundles for low-wage workers looking to migrate. We base these recommendations on community knowledge collected through surveys, on-ground organizations, and market/academic research. This knowledge is supplemented by processing proxies for certain influencing factors like using daily traffic density patterns to infer the safety of a neighborhood. We incorporate these elements, as well as anonymized live data from our platform, into a robust machine-learning algorithm that provides optimized employment and housing bundles for a worker based on stated and revealed preferences. 

The success of this effort requires a very agile and competent tech team which will require many iterations to develop a resilient and responsive algorithm. We would use this prize to hire this tech team in-house.

Currently, there are no public and transparent systems for determining the quality of a worker, employer, or informal housing in this low-income labor sector. Each stakeholder in the system has a perception of the situation which he uses to make his decisions, and these individual perceptions are shared with others in a very limited, unstructured way. This behavior results in the information asymmetries and inefficiencies that keep communities from having true control over their futures.

Bandhu’s platform will continually incorporate feedback from each stakeholder about the others through an anonymized rating system that helps share community knowledge and information. For example, if many workers flag a certain employer as exploitative, our algorithm accounts for this perception and pushes the employer to either change his practices or exit the platform.

Our solution ultimately gives migrant workers, traditionally beholden to their own limited social networks and exploitative systems, greater agency in deciding what jobs and housing will benefit them most. Work training programs through NGO partnerships will increase the worker's value to employers and give them a leg up in the urbanization process, accelerating inter-generational socioeconomic mobility. It also helps workers become the drivers of their own future by encouraging them to save.

We would use this prize money to give low-income workers performance incentives and invest in their up-skilling and training.

In most of the developing world, women—particularly single women—face many barriers to greater agency. They may want to pursue a career in a city, but are blocked by patriarchal expectations and legitimate fears over their safety.  Several women we interviewed in India who were widowed lived in destitution in villages, unable to work in urban jobs like many of the men of their age who were of the same marital status. They risked being ostracized from their communities and social circles if they made an independent move.

Bandhu matches women with the optimal opportunities so that they can migrate safely and with assurance. We do this by providing complete transparency and perspective on their critical next steps, allowing each woman to make the final decision on when and where she will work and live. Women seeking to migrate and contribute to their families’ income can use Bandhu to evaluate employer and housing ratings, particularly for those that have been rated by other women.

Usually employers pay women 20-30% lesser wage than men for the same job in this sector; we hope to use this prize funding for our pilot to fill this salary gap and put both genders on an equal financial playing field. The prize will also support female motivators in communities who do personal outreach and assist women access opportunities. In the long run, we would like to use prize funds to embed women-run enterprises in the job-housing-migrant system that uplifts siloed and home-based women.

Internally displaced refugees, whether driven by community persecution, climate disasters, or ethnic conflict, are one of the most vulnerable populations seeking stability and agency. After migrating to cities in search of work, these refugees are often connected to exploitative job and housing markets which disregard their safety, do not pay them enough or on time, and effectively lock them into indentured servitude through refugees’ inability to move to another location. The stress is further exacerbated by a need for the refugees to support a dependent family.

Bandhu offers refugees equitable access to job opportunities and low-cost housing without the fear of exploitation. Refugees’ agency, which is amplified through their own word-of-mouth networks, is furthermore made accessible to the larger community via digitization. Our algorithm would show refugees anonymized ratings that people in similar circumstances have made, often capturing critical community knowledge that chain migration networks rely on. Thus, local housing and job resources are pooled and vetted to create a rich database of community knowledge that can be relied upon.

We would use this prize to put additional human resources on-ground to better capture this population’s perspective and inform our algorithms, knowing well that data points and ratings may not always capture the mindset of this group adequately. We will use the prize to mobilize workers on the ground through NGOs like Aasmaan, conduct immersive workshops and focus groups with refugees, homeowners and employers.

Some of the biggest challenges today involve anticipating the migrations of the future being driven by international/domestic conflict and climate change. The world is becoming smaller and easier to navigate, but what is missing is the individual’s ability to make a truly informed migration decision. This leads to chaotic and piecemeal migration patterns that do not help a family’s ability to build a work reputation, save money, or send its children to school from a stable and safe home. Bandhu fills this gap innovatively by providing transparency to all on the platform and redirecting current informal social networks towards more equitable ends, while improving the socioeconomic conditions of the individuals within those networks. It does all this without the migrants having to compromise on community ties or their social capital.

We would use this prize to incentivize community leaders and middlemen to engage in more helpful behavior towards their colleagues and workers in their upward quest for socioeconomic mobility. The process would be supported by data analytics showing which workers under which middlemen show the highest increase in employer rating and net wages over time. These metrics could be used to reward the networks creating maximum positive impact for their community members. Incentives for helpful behavior will include bonuses, premium member statuses and greater opportunity to expand one’s business networks. We seek not to disrupt, but to create just transitions through nudging and realignment of actors and incentives.