The Centuries-Old Financial System Better Than DeFi
(Bloomberg Opinion) -- In the history of dangerous naivety, the decentralized finance mania of 2021 will hold its own against the 2007 boom in collateralized debt obligations. It took a financial crisis for the world to wise up to CDOs, which repackaged risky mortgage bonds to make them look safer than they were. “CDOs are nothing but a massive Ponzi scheme,” said the villain of a fictional account of the 2008 meltdown. How much more carnage will it take this time to know that blockchain-based lending is similarly reckless?
The idea that one could ditch regulated intermediaries like banks and make far higher returns by lending digital assets was a key attraction of decentralized finance, or DeFi. But that was before the bloodletting began, triggered by the collapse last month of the cryptocurrency pair Terra-Luna. The appeal of changing money into TerraUSD, a stablecoin that promised 1:1 convertibility into dollars, lay in the near-20% yield on TerraUSD deposits. Withdrawal of funds from Anchor Protocol, the main DeFi lending application on the blockchain, crushed the coin, as well as Luna, its sister asset.
Soon after, lenders Celsius Network and Babel froze deposits. BlockFi Inc., a Peter Thiel-backed lending platform, said it “fully liquidated or hedged all the associated collateral” of a large client believed to be Singapore-based Three Arrows Capital, a troubled crypto hedge fund. BlockFi is reducing headcount by 20% just as Coinbase Global Inc., the largest US-based digital asset exchange, lays off 18% of its workforce. There’s no end in sight to the crypto winter. Of the $252 billion of investor funds tied up in DeFi protocols last December, less than $75 billion remain.
Blockchain technology promised the Impossible Burger version of finance: lending without trust, the most important ingredient. Market participants in DeFi are anonymous. “Assessing the risk of borrowers through time-tested methods — from banks’ screening to reliance on reputation in informal networks — is therefore not possible,” researchers at the Bank for International Settlements noted recently. Thus, loans have to be over-collateralized in order to make up for the missing trust. But as recent events have shown, Bitcoin loans with Ethereum collateral can be just as combustible as the portfolio of subprime mortgage bonds backing a CDO.
Contrast the fragility of Defi with the robustness of “hawala,” a highly efficient system of moving funds in the Middle East and the Indian sub-continent since medieval times. If DeFi relies on software code to act as a substitute for courts in enforcing contracts, hawala seeks to fill the legal void with trust. As Matthias Schramm and Markus Taube described the institutional arrangement in their 2003 paper:
“(Hawala) is able to move large amounts of money without recourse of the formal banking system and even without retaining any bookkeeping notes. Instead, it is based on the trust of the participating parties and its social and religious embeddedness within the Islamic community.”
Modern-day regulators detest hawala because users of the multinational, club-like network can circumvent anti-money-laundering and terror financing laws with ease. Yet, the way the system operates, it’s almost impossible to obliterate, or even detect. Hawala intermediaries often maintain regular banking relationships indistinguishable from legitimate small business accounts.
Good or bad, hawala is a very real money-transfer product — and has been for centuries(1). By contrast, much of DeFi is just decentralization kabuki. Crypto bros talk big about defying the tyranny of government controls and large custodial organizations, though in reality DeFi can’t even match the success, in this regard, of a premodern alternative. Hawala arose to get around the lawlessness that preyed on medieval traders traveling long distances; it then learned to live outside — but alongside — the law.
That isn’t all. To be a DeFi borrower, you need more crypto collateral than the loan you’re seeking. This restricts “access to credit to borrowers who are already asset-rich,” the BIS report notes. For DeFi lending to become a serious tool of financial inclusion, two things have to happen. First, people need to be able to take loans under their real names to establish a pattern of trustworthy behavior. Second, more real-world assets like buildings and equipment must get digital representations on the blockchain so that even the less wealthy have some initial collateral.
For all the concerns around large tech platforms profiting from consumer data, fintech is doing a lot better than DeFi on inclusion. Online commerce platform MercadoLibre Inc.’s machine learning-based scoring model is demonstrably superior to what credit bureaus can tell a conventional bank about borrowers’ creditworthiness in Argentina. Ditto for Ant Group Co.’s Alipay payments network in China. Fintech has added a wider range of information — about a broader set of potential borrowers — to what traditional lenders could find out about a narrow group of people within existing banking relationships. This has had a large impact on emerging markets. A jar of Nutella sold by a mom-and-pop shop in India now tells a potential lender something valuable about its owner’s creditworthiness.
Ignoring borrower-level information — or losing it in the labyrinths of financial engineering — doesn’t end well. Think of highly rated senior CDO tranches where the underlying mortgages were subprime. DeFi needs to give up on its techno-anarchist utopia, and get more real and centralized. Otherwise DeFi lending will enter the annals of finance as a failure where hawala has been success: a 21st-century trustless technology losing to a 14th-century innovation that thrived by holding trust supreme.
More From This Writer at Bloomberg Opinion:
- When Crypto's Own Hedge Fund Geniuses Failed: Lionel Laurent
- Dreams of an Algorithmic Stablecoin Won’t Die: Trung Phan
- Central Banks Can Save DeFi. Really: Andy Mukherjee
(1) Hawala as a legal concept came to be described as early as 1327, according to Schramm and Taube, though the actual practice is probably much older.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Andy Mukherjee is a Bloomberg Opinion columnist covering industrial companies and financial services in Asia. Previously, he worked for Reuters, the Straits Times and Bloomberg News.
More stories like this are available on bloomberg.com/opinion
©2022 Bloomberg L.P.