Many aspects of the data science work in supplier-led trade finance are similar to prospecting: instead of tons of gravel, the data scientist sifts through tons of data (SME credit applications, supplier accounting data, books of receivables, etc.) in search of valuable nuggets – i.e., the low-risk suppliers to lend to. The data scientist’s toolset is, of course, amazingly more complex and advanced than a gold pan, but the principle is the same.
But what if, at least when it comes to trade finance, the Philosopher’s Stone was more than a myth?
The majority of SMEs’ credit risk is high; that is why it takes so much data prospecting to locate those valuable, eligible financing candidates and to identify the lowest possible rate to offer. But what if it was actually possible to transform the SME’s risks by re-focusing the analysis? Instead of looking at the SME supplier’s risk, we analyze the buyer’s likelihood to settle a given invoice. In this case, the type and magnitude of the risk can change dramatically.
An invoice by a small, credit-challenged supplier may have a terrible supplier risk, but it could have a very high probability of being settled. If the buyer is a highly rated, multinational corporate buyer with significant cash flows and long, good credit history, the risk has suddenly become a very attractive risk to finance.
Instead of searching for small amounts of gold in a river of high risk lending opportunities, we can open up new financing options based on some of the world’s biggest and best credit-rated businesses. This requires, however, the ability to accurately measure the probability that a given invoice will be settled by the buyer.
Traditional buyer-driven Supply Chain Finance (SCF) programs have taken a first step in this approach by focusing on invoices that have already been approved by the buyer. Here, the settlement risk is zero by definition, but requiring an approval restricts the scope of the program significantly.
Approval-based SCF programs typically reach only a few hundred suppliers in the top percentile by size (suppliers that typically have easy access to cheap financing anyway and so do not need trade finance), and are practically not accessible to the thousands of SME suppliers in the tail of the supplier base (the ones that really need trade finance). There are a large number of suppliers, and therefore financing opportunities, that this approach leaves untapped.
So, reliance on buyer approval cannot be the solution. Instead, we must assess the buyer settlement risk of an invoice without limitations to scale and scope. In other words, every single invoice as soon as it is issued.
This is possible by leveraging the wealth of data that the ERP systems of large corporate buyers can provide. By applying state-of-the art machine learning methods to this data, we can train a scoring algorithm that produces a “probability of settlement by the buyer” for any invoice. The nature of the supplier does enter the score, but ultimately what it measures is the quality of the invoice.
Hence, the advent of modern machine learning technology has made possible in the field of SME financing what medieval alchemists could only dream of: The transformation of leaden SME risk into golden investment-grade multinational corporate credit risk. But this transmutation requires the application of a technology that would have been inconceivable to a medieval alchemist.