Early this year I did a deep dive into crypto. During my exploration, I started to jot down some ideas that I thought were interesting. One of the things I really like about crypto is that it provides an exciting new lens to rethink the way we do things on the internet, especially in financial services. I also like that every participant in crypto networks has access to the same set of information. So in the spirit of open-sourcing everything, here's a first half-baked idea for how credit risk experts can band together to transform the world of consumer credit.
The underwriting processes utilized to assess and price risk for consumer loans are unnecessarily fragmented and opaque. Across the globe, consumers apply for credit at financial institutions without any understanding of how those creditors make a determination as to whether they will approve a loan, or how a loan is priced and termed. Different nations and institutions have a variety of disparate mechanisms and data sources that are used to make underwriting decisions. These range from FICO scores to assessing cash flow from a bank account and using assets like homes, land, vehicles, and livestock as collateral. Underwriting is opaque, fragmented, and inefficient. Tens of thousands of institutions are solving the same problem in similar ways, with little to no effort focused on collaborating to create a more efficient market that empowers more consumers.
Blockchain infrastructure enables a paradigm shift: instead of underwriting - the “proprietary” rules that power consumer lending - being controlled by siloed financial institutions that capture the majority of the value in this trillion dollar market, a free and transparent underwriting protocol that powers consumer lending at a global scale can be freely used by lenders, borrowers, and application developers, eliminating the need for centralized, gatekeeping financial institutions.
There are several different constituents that are required to create a globally decentralized and scaled consumer lending ecosystem: credit risk experts (underwriters), investors (lenders), borrowers (consumers), and application developers (distributors).
Professional credit risk experts and data scientists are the key to developing, maintaining, and evolving underwriting protocols that adapt to the ever changing global financial system. Currently, credit risk experts reside inside siloed institutions like online lenders, banks and credit card issuers. Many work on similar things, but don’t benefit from shared learnings, data or collaboration. They must all independently develop and test their models by trusting others in their institution to attract lending capital and borrowers in order to put them to the test. A better alternative is for all credit risk experts to band together to create a global set of underwriting systems that can cater to every consumer in the world regardless of geography or demographics (e.g. pseudonymous and crypto-only). Participating in open-source underwriting at global scale is the most effective way to create a transparent and efficient market that connects borrowers and lenders.
For investors seeking yield today, they must trust the mysterious underwriting methodologies of financial institutions if they want exposure to the consumer loans asset class. Large institutional investors can conduct diligence on these processes before making an investment, but this requires a cumbersome and bespoke process and information that only represents a snapshot of a particular moment in time. A better solution for institutional investors seeking exposure to this trillion dollar market is to have constant unfettered access to a completely transparent set of underwriting policies whose rules are set by smart contracts on a blockchain. This solves the problem of having to trust the underwriting criteria provided by a single centralized financial institution. A set of global underwriting policies hosted in a smart contract also enables institutional investors to access global scale through one protocol instead of having to identify many different institutions spanning multiple geographies.
For consumers looking for a loan (e.g. to refinance credit card debt, to finance a large transaction, etc.) it can be difficult to find a lender that will say yes, and to confidently know they are getting a good deal. Most lenders give consumers little choice in regards to their options. If they will lend, there are few toggles for the consumer to adjust interest rate, amount, or the term of the loan. Consumers do not know what lenders are specifically looking for in order for them to say “Yes” - whether it’s their FICO score, annual income, assets they own, or otherwise. Furthermore, they don’t know how a loan is priced, or what data points will lead to a larger or cheaper loan. It makes no sense that a consumer can’t understand exactly why or why not they can get a loan, especially after they are denied by a financial institution. This lack of transparency forces consumers into bad products. This should be a process that empowers consumers, both economically and educationally, not one that leaves them disenfranchised from the financial system.
For application developers who want to embed lending as a service offering, they can simply use the Protocol to offer credit to their customers. Whether it’s buy now pay later solutions, mortgage offerings, or the desire to use lending to strengthen their value proposition, they can now participate in a market without having to become experts themselves. Established financial institutions and lenders won’t need to exist in this new world as their offerings become commoditized and democratized by the Protocol, and the value they capture is transferred to all network participants: underwriters, lenders, borrowers, and distributors.
Many lenders specialize in one specific type of underwriting (e.g. prime consumer loans, subprime auto-loans, etc.). Smart contracts enable us to build a system where the protocol can host infinite types of global underwriting models ranging from merchant financing to mortgages and buy now pay later, updated and maintained by leading experts from across the globe. Oracles enable these models to use off-chain data and on-chain data to cater to different types of borrowers and lenders. The combinations are seemingly infinite, and all models are transparent by default to all parties.
The most critical element is getting a critical mass of credit risk experts to contribute to the protocol across the globe. To mitigate this risk, the optimal structure for this Protocol is a DAO, where top contributors are rewarded with tokens for their efforts in creating, maintaining, and evolving the various underwriting structures within the protocol. This must be a community-governed protocol. The “leading expert” is a global community/network of leading credit risk and data science specialists. Token distribution should account for all constituents in the protocol: underwriting contributors, borrowers, lenders, distribution partners, and the core maintenance team and its investors.
There are a lot of holes in this. How is recourse for defaults managed? How is it possible to operate across the global patchwork of byzantine financial regulations? There's plenty wrong with this idea, but there's also probably a kernel or two in here that's right enough to make something work in the world of consumer credit. Here's hoping that happens one day in the not too distant future.
Thank you Morgan Beller and Greg Rosen for helping me navigate the world of crypto, and Frank Rotman and David Snitkof for your feedback.