Decentralized Finance: What It Is, Why It Matters

There’s been a lot of hype, buzz, skepticism, confusion, and excitement around decentralized finance aka “DeFi,” the ecosystem of blockchain-enabled products and services that replace traditional financial intermediaries with freely accessible, autonomous, and transparent software.

While it’s still early days for DeFi (just a couple years in), its associated economy is already large and consequential: Ethereum, the backend infrastructure for DeFi, settled about $1.5 trillion in transactions last quarter, or 50% of Visa’s payment volume; decentralized money markets are issuing billions of dollars worth of loans every month; and individuals and businesses are using platforms like the Uniswap protocol to trade volumes roughly 30% the size of Coinbase’s. [Disclosure: I work for Uniswap Labs, which helped invent the protocol.]

Since there’s so much interest in this trend from entrepreneurs, corporate leaders, policymakers, and institutions big and small, I will try to explain the features and benefits of DeFi, outline some challenges ahead, and consider the road to mainstream acceptance and adoption.

But first, what made it all possible?

DeFi builds on three major waves of blockchain innovation across the last decade, each of which began with deep skepticism and has since progressed to acceptance and adoption.

The first era was defined by Bitcoin (invented in 2009), which gave us the distributed ledger, or blockchain, designed to facilitate peer-to-peer transfers of a non-sovereign digital asset. The second wave was defined by Ethereum, which drew from the same underlying distributed, censorship-resistant architecture: however, unlike Bitcoin, Ethereum’s native programming language (Solidity) can be used to create any conceivable application, transforming it into a globally accessible supercomputer. The third wave was the initial coin offering boom of 2017, which financed a range of projects, some of which have started delivering on their promise of a decentralized financial ecosystem.

DeFi is the fourth wave, and it builds on a combination of these innovations.

With DeFi, anyone in the world can lend, borrow, send, or trade blockchain-based assets using easily downloadable wallets without having to use a bank or broker. If they wish, they can explore even more advanced financial activities —  leveraged trading, structured products, synthetic assets, insurance underwriting, market making —  while always retaining complete control over their assets.

DeFi protocols abide by key criteria — in particular, permissionless-ness and transparency — reflecting values found in Ethereum, the open-source decentralized software platform that forms the infrastructure for most decentralized applications.

“Permissionless” speaks to both end consumers and developers: DeFi applications can serve anyone in the world with an internet connection, regardless of ethnicity, gender, age, wealth, or political affiliation. Additionally, any set of developers can confidently build upon these platforms, knowing that no central authority has the ability to revoke access in the future.

“Transparent” refers to the inherently auditable nature of DeFi platforms: Because the software is always source-available or open source, all underlying code is perpetually available for review, and all associated capital is open for audit. The transactions are all recorded on a blockchain, which enables easy review of particular transactions or building businesses exploring the data for investment (or even investigative) purposes.

DeFi’s two foundational qualities — permissionless-ness and transparency — translate to multiple, powerful use cases.

The permissionless nature of Ethereum-based applications — with the ability to freely and seamlessly “fork” (or copy and adapt) codebases — collapses barriers to entry for entrepreneurs down to zero. End consumers are the primary beneficiaries of this innovative environment: Because all applications share the same database (the Ethereum blockchain), moving capital between platforms is trivial. This forces projects to ruthlessly compete on fees and user experience.

A relevant example here is the rise of “exchange aggregator” applications: Using public APIs, these aggregators tap into multiple liquidity venues, splitting orders across platforms to provide end users with the best possible exchange rate. In just a matter of months, such aggregators have accelerated DeFi’s journey to best execution, a standard which required formal regulation for early electronic markets to converge upon.

Contrast DeFi’s competitive markets with consumer banking as it exists today, where opening and closing accounts can take three days. Or compare DeFi to the brokerage business, where transferring securities between different platforms takes up to six business days and numerous phone calls. Coupled with other onerous terms, these are the “switching costs” that discourage consumers from taking their business elsewhere, even with the pain of inferior services. Indeed, to the detriment of retail consumers, traditional finance is moving in the diametrically opposite direction, with the number of bank charters declining at an annual rate of 3.6% since 1990, limiting consumer choices.

The auditable nature of capital reserves in DeFi allows for rigorous risk assessment and risk management. For decentralized money markets and credit facilities — repo-like platforms that allow users to enter into variable-dated, peer-to-peer, secured-lending arrangements — users can inspect both the quality of the collateral portfolio, and the extent of leverage in the system at any given time.

Contrast this with the opaque nature of the current financial system. It was only after the global financial crisis of 2007-2008 that analysts and regulators began to understand that the ratio of loans to deposits in the U.S. had hit 3.5… double the ratio in the second most highly leveraged banking system, Russia.

The use of trustless, programmable escrow accounts (commonly known as “smart contracts”) allows DeFi protocols to bake in recourse, at the protocol level.

For example, in the MakerDAO system (a decentralized credit facility), MKR token holders earn interest paid by borrowers. However, in the event of insolvency or defaults, they serve as the primary backstop: MKR is automatically printed and sold to the market to cover losses. This programmatic enforcement creates very strict accountability, forcing MKR holders to set sensible collateral and liquidation risk parameters. The alternative — loose risk management practices — puts MKR holders at risk of dilution.

Compare this to traditional finance, where shareholders directly lose out when management makes mistakes. Archegos’ recent collapse serves as a recent example: While several senior executives at Credit Suisse left the bank, they were not personally held liable for the losses. With DeFi, however, direct accountability would translate to better risk management.

Ideally, capital should be as seamless as information in the internet age. In particular, settlement should be instantaneous, transaction costs should be minimal, and services should be accessible 24/7/365: 24 hours a day, 7 days a week, 365 days a year. It is simply not productive for our global financial system to only operate between the hours of 9 to 5, barring weekends and holidays.

There is clearly latent demand for modernized settlement infrastructure, as shown by Ethereum settling $1.5 trillion in transaction volume last quarter, up from $31 billion in Q1 2019. We also recently saw the types of market dislocations that can emerge from the absence of instantaneous settlement: Robinhood was briefly forced to pause buy orders for GameStop due to its difficulty in keeping up with capital requirements, which itself is a byproduct of T+2 settlement (the industry standard in which transactions typically take two days to clear).  

Efficient markets also require robust infrastructure. The distributed nature of blockchains gives them incredible resilience: In the six years since the launch of Ethereum, the network (and by extension, the applications built on top of it) boasts 100% uptime. The same cannot be said for centralized analogues. Even when centralized, established, and/or regulated, these centralized entities — whether exchanges or payment networks — can demonstrate unreliability, especially during high-volatility periods.

The consequences for consumers are very real. Take the example of brokerage users later logging on to find that their balances have rapidly diminished.

Inherently international markets have access to a larger pool of liquidity, significantly reducing transaction costs for all market participants.

Today, decentralized exchanges can offer better exchange rates for certain assets than siloed centralized exchanges or service providers. In equities markets, instruments like American Depositary Receipts (ADRs) serve to bridge access to foreign exchanges, yet often suffer from significant premiums and thin liquidity.

When markets are globally accessible, it can lead to more financial enfranchisement as well. Currently, developing nations are often excluded from financial services due to the high cost of setting up local operations relative to demand, lack of infrastructure, and more. But decentralized financial services — internet-native services with zero-marginal-user costs — can serve marginalized demographics, providing access to services like insurance, international payments, dollar-denominated savings accounts, and credit.

A by-product of building financial services on a transparent, shared database is that all associated transaction data is publicly available in real time. For example, earnings generated by liquidity providers in the Uniswap Protocol can be tracked on per-second granularity. Investors can use this data to decide how to allocate capital, providing for more efficient price discovery and allocation of resources, while regulators can monitor real-time transaction data to identify nefarious user activity.

This is a significant departure from traditional capital markets, where investors are left entirely in the dark until firms issue their quarterly earnings reports. The state of private markets is even more dire, with companies often inventing their own accounting metrics, if they decide to release metrics at all. It is difficult to imagine investors making rational decisions when having to work with stale data! Regulators also struggle in the current system, waiting years to discover wrongdoing, by which time it is often too late to rectify; Greensill Capital and Wirecard serve as two recent case studies.

By definition, DeFi platforms are “self-custodial”: users never relinquish custody of their assets to a centralized operator. While perhaps initially daunting for some people, the self-custodial nature of DeFi serves to eliminate counterparty and credit risk — the risk associated with a party in a financial transaction defaulting or failing to meet their obligations on a trade or loan. Analysts estimate that over $7B worth of cryptocurrencies has been lost through centralized exchanges since 2011, whether due to hacks or operators purposefully absconding with user funds. DeFi is a paradigm change, shifting “don’t be evil” to “can’t be evil.”

Self-custody is equally advantageous for operators, who can recuse themselves from unnecessary liability and compliance overheads: As an example, FinCen’s cryptocurrency guidance requires companies that custody user funds to acquire money transmitter licenses, a typically arduous process, while those that interact with self-custodied wallets can operate without.

As with any new and evolving technology, there are challenges ahead for DeFi. This is not unlike the early days of the internet, where connections were slow, hardware was expensive, and even the brightest innovators struggled to support the idea of images or video, which are now the very currency of online social activity.

The underlying backend infrastructure for DeFi, Ethereum, must continue to scale in order to support higher bandwidth demands. Processing approximately 1.5 million unique transactions per day, Ethereum is already at its current max capacity, and transaction fees have spiked as a result.

However, scaling must not come at the expense of security and decentralization. After years of deep R&D, multiple scaling solutions are now on the cusp of going live, promising to alleviate Ethereum’s burden while maintaining its core value set. Scaling will likely always be a fairly incremental process, with new capacity spun up to support excess demand on an ad hoc basis. (This, too, is not unlike the evolution of the internet.)

The DeFi onboarding experience is still too overwhelming for the average user. The process of moving fiat money (dollars, euros, sterling, etc.) into the crypto economy remains full of friction, with fiat on-ramps still limited to specific geographies and processor fees uncompetitively high.

Even after fiat has been transformed into crypto assets, custody and wallet management can be intimidating: Specialized “wallets” must be installed to interact directly with the Ethereum network — many requiring users to secure highly sensitive passwords, private keys, and seed phrases — without the comfort of “forgot your password?” backups. There’s no recourse in the event of misplacement.

There is reason to be optimistic, however. The industry is trending towards best practices on the custody and wallet front: For instance, “smart wallet” Argent avoids seed phrases entirely and provides users with both daily spending limits and a seamless “social” means of recovery in the event that devices are misplaced. I expect the fiat on-ramp business to become more competitive over time, with fees, coverage, and processing times improving as a result.

Global regulators have a lot on their plate as technology upends new markets. In the financial sector alone, regulators today are dealing with various forms of fintech ranging from neobanks and crowdlending to gamified stock trading.

Blockchain technology is an area that the traditional financial world, let alone regulators, overlooked or dismissed for years. Now, those regulators are evaluating the technology, the markets, and the participants to decide on the appropriate rules. Their goals are to ensure sufficient transparency for users and law enforcement (where transparency does not already exist); to target fraudulent behavior (while abandoning their earlier assumption that all blockchain-based activity was fraudulent); and to protect freedom of expression and privacy for consumers.

Yet many policymakers and regulatory agencies over the years proposed regulating cryptocurrency in ways that would have stifled each of the previous cryptocurrency waves, despite the potentially pro-consumer benefits. Instead they focused on the negative — illicit finance with bitcoin, high-risk investments with Ethereum and early token sales — often without recognizing the far greater positives.

As a result, there have been regulatory proposals that misunderstand DeFi — both the role of different actors and of the technology — and that would impose liability and burdens far beyond current law, and on largely uninvolved software developers. These proposals are akin to trying to hold the inventor of SMTP responsible for every spam email ever sent, or holding the inventor of HTTP responsible for every illegal website.

In much the same way “private blockchains” were the product of incumbents mistakenly seeking to co-opt earlier blockchain waves, there is a risk that centralized financial institutions co-opt the DeFi movement, making considerable concessions along the way. Although seemingly identical to other smart contract blockchains like Ethereum, some of these chains are practically centralized, betting that users focus on speed and low fees — while sacrificing the permissionless, neutral, and immutable guarantees that sit at the heart of the DeFi value proposition. While there may certainly be “weak” forms of tech that arrive with every tech trend alongside their strong forms (as Chris Dixon has argued), this is not a case of weak forms of said tech but rather misleading marketing, an old wolf in a new sheep’s clothing.

It is not difficult to imagine some traditional financial institutions like commercial banks, major tech companies, or even nation-states who instead of figuring out how to adopt or integrate DeFi, do the misleading or less strong version instead. And while these latter offerings may unlock incremental efficiency gains, they won’t meet the full potential this technology can offer: global, permissionless access to global liquidity and the total elimination of counterparty risk.

* * *

DeFi is here, and it’s here to stay. Some skeptics think of it as an idealistic movement, destined to forever be relegated to the shadows of the internet. But due to its novel innovations around settlement efficiency, risk management, and accessibility, DeFi is likely to become a central piece of the financial infrastructure not only for cryptocurrencies, but also potentially for all other classes of markets: In the not-too-distant future, people will sell tickets, Apple stock, pork belly futures, socks, and much more using DeFi protocols, likely with portals providing access to that infrastructure with separate regulatory regimes and business operations.

It won’t spell the end of the existing financial services sector, as some hardcore believers may argue (just like the internet did not altogether kill print). But the opportunity in DeFi for traditional financial services and other companies here will be in allowing them to focus on their core structural advantages — custody products, prime brokerage, fiat on-ramps, customer service, and so on — while sourcing liquidity and products directly from decentralized protocols.

Early skeptics had said that nobody would use or value bitcoin; in just over a decade, it has become a trillion-dollar asset rivaling gold, and is held on the balance sheets of several public companies. Similarly, detractors argued that Ethereum would not work, was too slow, and was too expensive. Ethereum today supports thousands of permissionless applications, has settled trillions of dollars worth of transactions, served as infrastructure for legacy financial giants, and has contributed enormously to cutting edge cryptography research. And even with the failures of the ICO boom, many token sales funded the development of extraordinarily important technology, including decentralized storage (a longtime holy grail in computing), network interoperability, manipulation-resistant data feeds, and decentralized computing.

Most importantly: each of these waves of crypto attracted tens of thousands of engineers and entrepreneurs, which is exactly how the future of DeFi will be built.