Why Did Most DeFi Protocols Disappear in 2021?

Written by: Vaidik Mandloi

Compiled by: Chopper, Foresight News

Between 2020 and 2022, a group of astute crypto investors bet on a number of protocols that built on-chain fixed-rate lending products, allowing users to lock in yields similar to government bonds. Element.fi raised $32 million in funding from a16z crypto and Polychain Capital, while Notional.finance secured $10 million in a round led by Pantera Capital, followed by at least five other similar projects.

Today, without exception, these projects have all disappeared. The yields they attempted to package into products were entirely fabricated, primarily because the DeFi of 2021 relied on protocol-generated token rewards and leveraged cycles to create returns. Once users concentrated their exits, the entire system collapsed.

In this round of industry cleansing, the only surviving protocol is Pendle.finance. The reason it survived is that treasury assets on RWA.xyz began to go on-chain, bringing real cash flow, allowing developers to build products around real yields. I believe a similar transformation is about to impact all existing DeFi foundational components.

Various Workarounds

To understand this transformation, we first need to grasp the market environment at the inception of DeFi.

Looking back at the on-chain financial market of 2018 and 2019, the tradable assets were all crypto-native tokens, with no cash flow, no legal entities behind them, and extreme price volatility, where values could drop by 90% in an afternoon.

Market participants were anonymous wallet addresses, with no credit records or legal identities, and in the event of default, there was no accountability mechanism. Ethereum could only process 7-8 transactions per second at that time, with block intervals of 13-15 seconds. The main participants were retail speculators, along with a few crypto-native funds willing to take on extreme risks for high returns.

The foundational conditions required for a functioning financial market were entirely absent in the on-chain market at that time. A mature market needs market makers willing to place bilateral orders and provide continuous quotes, the price movements of the underlying assets must be predictable to hedge risks; it also requires lenders to assess borrowers' repayment capabilities, which necessitates knowing the true identities of borrowers.

DeFi at that time fell far short of these standards. Asset volatility was too high, and professional market makers were unwilling to participate; borrowers were anonymous wallets, and the user base showed no interest in financial products with fixed maturity dates.

Thus, developers at the time began designing various workaround solutions. For example, Uniswap invented the automated market maker (AMM) not because it was a better trading method, but because no one was willing to make a market for tokens that could plummet by 90% in just a few minutes. The solution at that time was to write the pricing logic directly into smart contracts, allowing anyone to inject funds into the liquidity pool, with the constant product formula automatically completing price discovery, and each transaction adjusting the exchange price based on the token ratios within the pool.

Many DeFi native users still do not fully understand: this mechanism is fundamentally different from traditional market making. Traditional market makers earn the bid-ask spread. However, what AMM liquidity providers (LPs) do is entirely different. Whenever arbitrageurs find a price discrepancy between the liquidity pool and the external market and trade to close the gap, the LP on the other side is effectively paying the arbitrageur, completing the rebalancing of their own portfolio.

For example: you deposit an equivalent value of ETH and USDC into the Uniswap pool. Suppose ETH rises by 20% on Binance, but the liquidity pool has not yet sensed this price change. The arbitrageur will notice the undervalued ETH in the pool, buy it, and then sell it on Binance for a profit. Inside the liquidity pool, USDC increases, and ETH decreases, which is equivalent to you selling ETH during the price rise.

When ETH falls, the reverse operation occurs: the arbitrageur dumps ETH into the pool and withdraws USDC. With each transaction, your asset portfolio automatically returns to a 50/50 ratio, and the arbitrageur takes the price difference as profit. The underlying mathematical logic is interesting; even under extreme assumptions, if every transaction in the liquidity pool is purely arbitrage with no retail trading, after collecting small fees from each transaction, liquidity providers can still achieve positive returns.

Aave and Compound in the lending space also follow this design logic under constrained conditions. Traditional banks rely on credit scores and income records to assess your repayment ability when lending. However, on-chain anonymous wallets cannot do this. Therefore, these protocols adopted an over-collateralized lending model.

Perpetual contracts replaced traditional fixed-term futures for the same reason. Quarterly futures contracts require counterparties to fulfill obligations on fixed delivery dates. However, in the crypto market, anonymous funds rotate between various protocols in pursuit of the highest yields, making such long-term commitment to fulfill obligations impossible to guarantee.

Of course, these makeshift solutions have produced usable products. But from a broader perspective, the financial system they collectively built has a glaring issue: the collateral supporting all products does not generate any yield.

We can apply Hyman Minsky's theory to stress-test any financial arrangement: examine whether the income generated by the borrower's collateral is sufficient to cover the debt. Financial relationships can be divided into three categories.

The first category is hedged financing, the safest: asset income covers both interest and principal. For example, a salaried worker repaying a mortgage with their salary is a typical example.

The second category is speculative financing, an intermediate state: income can cover interest, but when the principal matures, refinancing is needed. Rolling issuance of corporate bonds by companies falls into this category.

The lowest tier is Ponzi financing: income cannot cover either interest or principal. The only hope for maintaining payments is that the price of the held assets continues to rise; once the price stops rising, the entire system collapses.

Now, looking at the scenario of a user borrowing on Aave, the borrower deposits ETH worth $10,000 and borrows $6,000 in USDC to freely use this stablecoin. However, the ETH in the Aave treasury does not generate any income to repay the loan. The only condition for maintaining the safety of this loan is that the market price of ETH remains above the liquidation threshold. Once the drop is significant enough, the protocol will liquidate the collateral and close the position.

The entire lending arrangement relies entirely on the asset price being able to hold until the borrower exits. Almost all DeFi over-collateralized lending using crypto-native tokens falls into this category.

How the Transformation Will Occur

If we replace the collateral in the scenario with tokenized U.S. Treasury bonds, the lending protocol and borrower remain the same, but the collateral is no longer ETH; instead, it is a tokenized U.S. Treasury bond worth $10,000 with an annual yield of 4.5%, allowing the borrower to borrow stablecoins at a cost of 3%. The collateral generates $450 in income each year, while the loan cost is only $300. The cash flow from the collateral can cover the debt cost. This yield is unrelated to the crypto market cycle; it comes from the coupon of U.S. government bonds, and regardless of whether the price of Bitcoin is $100,000 or $30,000, cash flow will continue to flow in. The longer the loan position is held, the more robust the borrower's position becomes. This sharply contrasts with ETH collateralized lending: with each market downturn, the borrower gets closer to liquidation, and the collateral itself does not generate any cash flow.

When the vast majority of collateral in DeFi yields zero and relies entirely on price appreciation, all positions are exposed to the same one-way risk. A drop in one type of asset triggers liquidation; the selling pressure from forced liquidations further depresses prices, leading to more liquidations and creating a chain reaction. Since 2020, we have witnessed this process in almost every crypto downturn cycle. People often regard it as a black swan event, but the fact that the collateral itself does not generate cash flow makes such an outcome entirely predictable. However, when a significant portion of the collateral consists of income-generating assets like U.S. Treasury bonds, even if the crypto market crashes, the positions remain stable because their cash flow is independent of the crypto market.

For the first time, the lending system has a safety net that does not rely on a bull market. The scale of tokenized money market fund shares has grown from $770 million at the end of 2023 to $14.82 billion on October 5, 2026, increasing nearly 19 times in less than three years.

This model also has precedents. In 1955, the Midland Bank in London began accepting dollar deposits. At that time, the U.S. Q Regulation limited the maximum interest rates on deposits by U.S. banks, while British banks were not subject to this constraint. Dollar funds flowed overseas in pursuit of higher yields, giving rise to a parallel dollar system, which reached a scale of approximately $4.7 trillion by the early 1980s.

However, the Federal Reserve did not suppress the Eurodollar market but chose to adapt to it. A parallel market born out of the limitations of the existing system will continue to expand until the original restrictions are lifted. Thus, the Q Regulation was gradually abolished, and money market funds emerged as a competitive product in the U.S. domestic market, ultimately integrating this parallel dollar system into the mainstream.

Stablecoins are the Eurodollars of this generation, while tokenized U.S. Treasury bonds are the money market funds of this generation.

The protocols that have disappeared attempted to build an interest rate market when there were no real interest rates on-chain, using token issuance and leveraged cycles to serve as real yields. But you cannot rely on yields that can be shut down at any time through protocol governance votes to construct a complete interest rate curve.

Pendle survived because it waited for the emergence of tokenized U.S. Treasury bonds and income-generating stablecoins, allowing the market to finally have real cash flow backed by the government to build products on. Pendle's Boros is now the first functioning on-chain interest rate swap product.

The same logic applies to the trading track. In trading pairs with good liquidity, AMM liquidity providers lose about 11% of their funds annually due to arbitrage losses; for tokens not listed on centralized exchanges, bearing this "arbitrage tax" was the only option. However, assets like U.S. Treasury bonds, stocks, and bonds, which have deep global liquidity, make pursuing the AMM route very unreasonable.

These assets require on-chain order books, with performance comparable to centralized exchanges. Major public chain teams have long claimed that this is technically unachievable, but Hyperliquid has overturned this conclusion. At the beginning of 2025, it briefly accounted for about 60% of on-chain perpetual contract trading volume.

All these signs indicate that the infrastructure built for tokenized U.S. Treasury bonds and tokenized stocks will be closer to traditional electronic trading markets, which may exceed many people's expectations, as these assets were originally service targets of traditional financial markets.

The first generation of DeFi foundational components was born under the conditions of anonymous wallets holding high-volatility, cash flow-less tokens. But now the underlying assets have changed, and the makeshift solutions of the past will gradually phase out. The new infrastructure that replaces them will converge with the traditional market architecture that has processed trillions of dollars in transactions daily for decades.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.