A DPI for Non-Financial Data

India's latest DPI focuses on leveraging non-financial data. By offering API-based access to a number of different data sources it will give lenders additional ways in which to evaluate the credit-worthiness of potential borrowers.

This is a link-enhanced version of an article that first appeared in the Mint. You can read the original here.

One of my most vivid childhood memories is of visiting the Amul factory in Anand, Gujarat. The father of India’s White Revolution, Verghese Kurien, was my grand-uncle, and as his guests, we were given a grand tour.

As a young child, the highlight of the weekend was being able to wander through the chocolate factory, surrounded by so much molten chocolate that it had to be transported from one machine to the other in overhead pipes. I vividly recall how one of the pipes had sprung a leak and marvelled at the fact that the factory manager had strategically placed a bucket below it to catch milk chocolate as it dripped down. This was the closest I was ever going to get to Willy Wonka.

That was also when I learnt about India’s milk revolution—directly from the mouths of those who were a part of it. I was told of how my grand-uncle, in defiance of a deeply entrenched milk mafia, helped the dairy producers of the district organise themselves into a cooperative so that they could sell milk directly to customers. And how, from those humble beginnings, dairy farming became one of the largest self-sustaining industries in rural parts of India. By offering dairy farmers a way to get directly involved in all aspects of the procurement, processing and marketing of milk, India was able to go from being a milk-starved nation to the world’s largest producer of milk by 2006.

But as much as the idea of a cooperative gave dairy farmers a way to access the market, a lot remains to be done to improve their lot. India’s dairy sector is deeply fragmented, with over 70% of the country’s milk being produced by small and marginal farmers, each of whom own just 1-2 cows or buffaloes. This means that their average net income is still relatively low. Unless we can find a way for them to scale their operations up, we will struggle to ensure the continued sustainability of the sector.

Key to this is access to capital. While most cooperatives do offer their members some form of loans and advances, dairy farmers typically lack the collateral required in order for them to avail a loan from the formal financial sector. This means that even though we have made it possible for them to collectively access the market, they still lack the capital needed to grow their individual business to sustainable levels. Given our demonstrable success at doing this, should it not be possible for us to build digital public infrastructure (DPI) to solve this problem?

The Reserve Bank of India recently unveiled the Unified Lending Interface (ULI), a new DPI platform designed to make available to lenders various types of non-financial data that they could use in order to evaluate the credit-worthiness of borrowers. In the context of dairy farmers of the Gujarat Cooperative Milk Marketing Federation, this could mean digitally verifiable data about the volume of milk contributed by each of them to the cooperative as well as its purity. The ULI makes it possible for lenders to access this information directly from records of the National Dairy Development Board using an API interface. This will give formal lenders new sources of data that they can use to evaluate the credit-worthiness of borrowers otherwise ineligible for loans.

This is just one of the many uses to which the ULI framework can be put. Lenders can now get access to a range of diverse data sources in an easily accessible digital format, making it possible for them to evaluate a much wider cross-section of potential borrowers than would otherwise have been possible. By overlaying multiple such data points (land record data, irrigation data, weather data, etc), they can arrive at a more fine-grained evaluation of the ability of potential borrowers to repay loans from one year to the next.

This, in many ways, is similar to what the account aggregator (AA) system has already made possible. Except that, while India’s AA ecosystem currently only enables data sharing between entities that are currently regulated by one or the other financial-service regulators, the ULI will provide lenders API-based access to non-financial data sourced from entities other than those regulated by these regulators. When combined, these two ecosystems will offer formal lenders a more complete and accurate picture than is currently available to them at a fraction of the cost.

So what are the data sources that could be made available through the ULI once the platform goes live? Quite frankly, the answer to that question is only limited by our imagination. So long as a data source is available at scale, it should be possible to serve it up using the United Lending Interface.

For instance, it should be possible for us to integrate goods and services tax (GST) data with the ULI, offering transaction-level information on sales. It should be just as easy to integrate customs information, so that we can offer traders new ways by which imports and exports can be financed. The same could be said of warehouse data, satellite data and even data from specific sectors such as coffee, where sales are intermediated by the Coffee Board.

Since it has been created using the same design philosophy that underpins all the other DPI in the country, the real value of the ULI lies its interoperability. Which means that once it is combined with other DPI platforms, and as long as we assure private innovation a free rein, we will probably see new products and services the likes of which are currently impossible for us to imagine.