DeFi, why did you go after standing up?

This is an inevitable trial and error. But where to go after wrestling is a question that the DeFi industry needs to think about and keep thinking about.


In July 2019, Deutsche Bank announced a major plan to lay off 18,000 people worldwide. At that time, a captured picture captured the attention of the major media. In this picture, two men in suits walked out of the bank's office building, one of them carrying a canvas bag with the words Bitcoin.

At the time of the decline of traditional banks, decentralized finance has also had such a slight subversion.

2019 is the year of decentralized finance.

Decentralized finance is the use of financial services based on blockchain technology. Products include decentralized loans, decentralized exchanges, and decentralized derivatives.

In the past year, blockchain technology has tried many landing scenarios. In the end, Ethereum abroad seems to have reached a certain consensus and believes that DeFi is the best application of the Ethereum smart contract.

It is estimated that the global bond market currently exceeds $100 trillion, the global stock market value exceeds 64 trillion, and the global derivatives market has a total market capitalization of approximately $12.7 trillion. If DeFi can get a piece of it, it will be a huge market.

The enthusiasm of DeFi is not without reason. In 2018, influenced by the bear market, decentralized lending represented by MakerDao gradually entered the public eye. Lending, stable currency and derivatives have become the way for people to survive in the overall decline in the overall asset price of cryptocurrencies. At that time, the DeFi project, including Veil, dYdX, and Dharma, successively received a counter-market investment from Wall Street Capital.

The market size has also grown rapidly. According to DefiPulse, on June 27th, the total lock-in position of the lending market hit a peak of about 602 million US dollars, which was about 129.77% higher than the 262 million US dollars on January 1 this year.

But in the near future, the fast-growing DeFi industry has seen two left behind. In July this year, Veil, the second-largest forecast market platform for Ethereum, was announced last week. The project, which was funded by well-known venture capital institutions such as Paradigm and Sequoia, survived for only six months. In early August, San Francisco's crypto loan agency Dharma also announced a decision to suspend new deposits and loans on its platform.

At the same time, people have found that the lock-up position of the entire lending market has been declining since hitting a high point on June 27 this year, and has dropped by about 30.59%.


So, there were some different sounds about DeFi. The founder of Primitive Ventures, Wan Hui, said on Weibo that DeFi is the concept of self-help in the Ethereum after the DApp concept was cooled in the ether.

Although it is too early to tell the story, we still have to throw a question: the new life-saving straw that is regarded as the blockchain after DApp is really strong enough to lead the blockchain, or the next cool one. What about the "DApp" concept? How long will this pain last before DeFi takes on the future of the blockchain?

Star project Veil and Dharma frustrated

In July, Veil, the star DeFi project launched only half a year, announced its suspension.

The predictive derivatives trading platform, jointly launched by the decentralized forecasting platform Augur and 0x, has been favored by Sequoia and the encryption fund 1confirmation and Paradigm.

At that time, DeFi's three major applications were lending, stable currency and decentralized exchanges. Veil, which aims to start the market with predictions, is seen as the power to revitalize the new DeFi track. In May of this year, Veil opened the 2020 US presidential election decentralized market forecasting service and once became a topic.

If you look back at Veil's daily activity and volume data before closing, it is not difficult to understand that Veil is difficult to pull. According to DefiPulse, since opening, Veil's total lock position does not exceed $90,000, which accounts for 0.01% of the total DeFi lock position. The parent company and the Aurgur lock position that focuses on the forecast market account for only 0.15%. . In the Veil page, the predicted event participation amount exceeds 10,000 DAI in addition to the presidential election. The vast majority of predicted events participate in less than 10 ETH.

Founder Paul Fletcher pointed out three of their dilemmas in the outage description:

1. We want to do too much. Forecasting the market is a broad form of gambling, derivatives and insurance. Centralized versions of these vertical fields may be more suitable for users than general forms;

2. We did not provide a good entry experience. The cryptocurrency is still at an early stage as a user base, and we have not provided enough convenience for users who do not have cryptocurrencies or wallets;

3. We are not decentralized or regulated. Some users want a completely decentralized, unstoppable product; others want a regulated product. It is difficult to provide something that people think is valuable between the two.

In general, it can be summarized that it is difficult to control the balance between predicting a small market, high user thresholds, decentralization and regulation.

For Veil's exit, Hydro Protocol & DDEX market public relations person in charge of Dai Shichao said: "No money, no secondary market burden, belong to the track trial and error stop loss, and also share the failure experience when exiting."

One month after Veil announced the shutdown, the Dharma platform on the decentralized lending platform appeared to be difficult for users to withdraw coins, the contract code was closed, and DappTotal's DeFi data showed that Dharma lock assets fell from $21.87 million a month ago to August. On the 2nd, 5.61 million US dollars, 74% of the assets have been transferred away from the currency, and the assets are out of the state, Dharma was once misunderstood as "running."

A week later, the founding team said that the suspension of the loan was to create a better user experience.

Dharma's background has many similarities with Veil. In February of this year, it announced the completion of $7 million in financing. The investment institutions include star capitals such as Green Visor Capital, Polychain Capital, Passport Capital, Y Combinator and Coinbase Ventures; the founding team is equally strong. The founder graduated from Stanford and worked for Google and the cryptocurrency exchange Coinbase.

When analyzing the reasons for Dharma's developmental resistance, the elimination of subsidies and P2P is a common voice. In February of this year, Dharma launched a subsidy policy for customers to reduce the loan interest rate; however, the subsidy was cancelled in May this year. According to data released in May, since the cancellation of subsidies, the total amount of locks has decreased by 60% compared with April.

The P2P matching mode is characteristic of Dharma. In the P2P matching mode, the borrowing and the lending parties are manually matched, and the lender is locked in the loan during the lending period. Only after matching with the borrower, the interest earning starts, and the borrowing rate is fixed. In contrast, Competitive Compound is a decentralized liquidity pool that allows users to share their assets with other borrowers on the platform to create dynamic interest rates based on the supply and demand of the asset pool.

Hydro Protocol & DDEX market public relations leader Dai Shichao also said, “Dharma is the P2P model. Whether the matching is successful or the speed of the currency is all depends on luck. The sudden drop in the lockout amount is related to their stop of interest rate subsidies, no interest rate advantage, plus Manual matching is inefficient, so users are lost."

She pointed out the third reason besides the P2P model and stopping subsidies, namely manual matching. The system does not have automatic clearing function, and needs manual settlement. The borrowing is approved by the centralized organization. The source of currency price information, interest rate information, and platform development and update are centrally controlled.

The dilemma of DeFi

With the development of the DeFi industry, the “dilemma” outside the blue ocean of the “opportunity” of the DeFi industry is gradually emerging.

User threshold and experience

In addition to the problems Veil and Dharma encounter, user thresholds and experiences are common to decentralized products, and DeFi is no exception. “Difficult to use” is a problem with too many DeFi products.

Hydro Protocol & DDEX COO Wang Bowen pointed out that the threshold for users is to manage their wallets, private keys, and understand contractual interactions. And these have requirements for financial and blockchain reserves.

Kiaber Network China market leader Lucas further pointed out that the current cross-chain technology is not mature enough, resulting in a lack of user experience in cross-chain transactions; in addition, the lack of performance of Ethereum will lead to network congestion, making the ether There is no way for DEX in the workshop to handle transactions well. For the user, this also creates an obstacle to the experience. The mobility of decentralized ecology is also different from that of centralized exchanges or applied ecology.

User thresholds and experiences directly contribute to the niche of users. At the StakingCon-Staking Eco-Conference, which was co-sponsored by Odaily Planet Daily and BlockBeats, PeckShield founder Jiang Xuxian analyzed the data of Makerdao and Compound and dYdX. The total borrowing of these lending platforms has greatly increased from this year. The increase from around 34 million in January to 220 million at the end of June has more than fivefold in just a few months. However, compared to the significant increase in the total amount of borrowing, the number of users of the DeFi project is relatively small. He said that this will be a major challenge for the entire lending industry.

Risk disclosure

At present, borrowing products can be divided into fixed-term and non-fixed-term loan options according to the loan term. The risk factors are different. For DeFi products, the necessary risk disclosure is also very important.

NUO, the untrusted lending agreement that has recently become a DeFi rookie, once hit the top five DeFi project rankings in April, and recently encountered user complaints due to risk disclosure.

A netizen named whuttheeperson posted on Reddit that the fixed-term loan pledge on NUO was liquidated within the agreed lock-in time. He believes that, unlike open-end crypto loans like Maker and Compound, the risk of liquidation of fixed loans is not large, and they have adequately adjusted the parameters to accommodate the risk of devaluation of risk collateral during that time.

He also believes that NUO has a huge oversight of risk disclosure in its loan process; on the contrary, the more mature MakerDAO will have relevant risk warnings at every step.

Encrypted asset limit

According to Yang Mindao, founder of dForce and Blockpower, the problem with DeFi is that there are too few cryptographic assets that can actually be mortgaged.

It must be acknowledged that even though DAI and USDC are absolute head-stabilized coins in the DeFi field, they are not truly universally adopted in China. The scale of de-centered lending is only 1% of centralized lending.

The current DeFi track has only a lot of players on Ethereum. Of the top 50 financial dApps, 42 are based on Ethereum, including MakerDAO and OmiseGO; 44 of the top 50 exchanges dApp are based on Ethereum, including Augur and Uniswap.

The single asset also makes the DeFi lock volume vulnerable to ETH market. For the recent decline in lock volume, Pan Chao, the head of MakerDao China, said that the main reason is that the ETH asset market is down, and many people change ETH to stabilize the currency. At present, the lock volume of the stable currency DAI has increased to more than 30 million US dollars.

Insufficient risk control system

Since risk control relies on mortgages, collateral prices affect risk. Thus, the chain of predictors (the mechanism for discovering and submitting real data to smart contracts) has become the focus of the current DeFi industry.

A price predictor that is accurate, fast, unbiased, and resistant to price manipulation is considered a “missing puzzle” for DeFi products. The current centralization and decentralized oracles have room for improvement: centralization means there is The risk of price manipulation, while the decentralized predictor still has insufficient incentives to provide real-time and accurate prices.

The denomination machine of the decentralized exchange Uniswap is considered to be the current leader. Instead of matching the buyers and sellers through the order book, it collects the liquidity of the market maker and uses algorithms to set the price. As long as there is a price difference, Will create an opportunity for arbitrage, the price will be quickly corrected. However, the current drawback of Uniswap is its poor liquidity. Some people call it “砸500 ETHs.”

Most of the current DeFi projects use the Makerdao oracle. According to Pan Chao, the Maker predictor is a distributed predictor, which selects the price of 14 exchanges and selects the median for the feed price.

Pan Chao believes that the problem of the prophecy depends mainly on what the pledge of the provider of the prophecy is, whether it is reputation or assets. To solve the problem of the prophecy machine, it is actually to solve the reputation problem in the chain; at present, such a reputation system is not mature in the chain.

The lack of a reputation system also affects the DeFi user experience.

"The biggest drawback of DeFi today is that it relies on over-collateralized debt positions." Fitzner Blockchain Consulting director Campbell said.

Taking MakerDAO and Compound in the forefront of the current lock volume as an example, currently, MakerDAO requires users to deposit 1.5 times the ETH price to establish a Collateralized Debt Position that supports Dai. Most people choose to keep their “loan and mortgage price” ratio at 300% to avoid double-digit liquidation penalties. In addition, Compound's loan interest reached 6-13%, and the collateral for all loans was 1.5 times the value of the loan.

Campbell said that the reason for the user's borrowing is in contradiction with the demand for excess collateralized capital. If the mortgaged asset price does not cover the position in time, the risk of liquidation may be faced.

Pan Chao believes that the current credit loan that DeFi industry lacks is based on a mature reputation system. At present, “the difficulty is still very large”.


The decentralized derivatives trading platform, dYdX, has highlighted three main commercialization models of the DeFi project: value-added through the project's native tokens, such as MakerDAO (MKR); revenue through fees, such as Compound; through user-oriented applications The program is commercialized. Such as dYdX, Dharma, etc.

However, the current problem is whether the DeFi project needs to be a native token. It is still a problem. Because the borrowing frequency is not better than the transaction, although the current loan interest rate is high, the loan is mostly short-term, the time value is small, and there is still room for improvement in daily living data. Fee income is not high. According to a report released on August 15th by the credit credit evaluation startup Graychain, in the past 18 months, approximately 244,000 loans were issued, with a total loan value of approximately $4.7 billion, but only 1.83% interest. Being recovered, how to ensure continuous and stable income has become one of the problems that many DeFi products must face.

Where is the development direction of DeFi?

Veil and Dharma's "success", in the eyes of most industry insiders, seems to be mostly due to their own factors.

Compared with poor management, the industry is more inclined to "reinvent the reconstruction". Hydro Protocol & DDEX COO Wang Bowen believes that Veil and Dharma are both changing directions, not closed. “DeFi's attempt is the same as that of a startup. If you find that the direction is wrong, there is no data growth, and you can change direction quickly.” He pointed out that Dharma's current suspension is to transform the previous P2P lending into a similar fund pool.

Yang Mindao also believes that it is not necessary to over-interpret. He said that Veil and Dharma are trial and error processes for very normal entrepreneurial projects, and the market should give them enough patience.

But Veil and Dharma are also showing that the DeFi on the wind is not a road to Rome, and there are different developments in different track markets.

Yang Mindao believes that the decentralized open financial agreement, the infrastructure provided to the B-end, has only begun to be built on a large scale, with plenty of opportunities. It is the correct posture to build a highly modular, detachable combination and interoperability.

In addition, infrastructure-level opportunities for tokenization of assets under the chain (such as stable currency is the chain of legal currency), is a market of 10 trillion US dollars.

Pan Chao also holds a similar view. He said that the stable currency is a necessity. To benchmark traditional finance, there must be a money market and then a capital market.

In addition, he believes that DeFi's future should be separated from the concept of “decentralized finance” and become open finance. If DeFi wants to expand ten times and 100 times the market, it is necessary to do traditional asset chaining, although the middle must be There will be trade-offs for decentralization, but if the business logic on the chain, such as hosting and mortgage risk settings, is sufficiently decentralized, it still meets the original intention of DeFi.

For example, he borrows DeFi applications, and the assets behind them are not completely decentralized assets such as Ethereum and Bitcoin, but synthetic traditional assets (traditional institutions endorsement chain), but all users of stable currency for mortgage lending have permission to use. This point of view still belongs to the category of DeFi.

Pan Chao said that he is also more optimistic about the bond market on the blockchain. The synthesis of traditional government bonds and corporate bonds, as well as the cryptocurrency's own bond products, from the perspective of the traditional money market, the best liquidity, is also the largest financial product. The recent staking of fire is the embryonic form of the bond market, provided that the system is constantly generating cash flow.

As for the restrictions on DeFi assets, Mindao believes that BTC on ETH (referring to the release of BTC-backed ERC20 tokens) is a potential project, “it will bring dozens of collateral assets and need a coin to solve the incentive problem. Alliances, exchanges, wallets, and project parties can all participate in BTC coinage, and personally feel that this alliance will have a breakthrough in the coming year."

Looking back at the birth of DeFi until now, but for a year.

This year was a year of rapid development of DeFi. The DeFi Ecology began to take shape in Ethereum, and applications such as stable coins, loans, derivatives, and decentralized exchanges were born.

But within a year, the closure of the star DeFi product has forced us to re-examine the DeFi on the wind. The future is still infinite, but it must be acknowledged that the responsibility of DeFi as the "blockchain future" is still unrestrained and long. Compared with the amount of funds in the traditional financial sector and the amount of users, today's decentralized finance is still insignificant, and users are almost exclusively from the inventory world of the encrypted world, and lending business accounts for 90%. These dilemmas stem from user thresholds, funding and technology limitations.

The good signal is that if you are poor, you will change your mind. The "difficulty" means DeFi's new "opportunity". Whether it is Coumpound, Uniswap and other products have been or will be launched V2 version, seeking evolutionary upgrades; or BlockFi, dForce and other DeFi products in the near future, have said that the industry still has a lot of imagination.

This is an inevitable trial and error. But where to go after wrestling is a question that the DeFi industry needs to think about and keep thinking about.

Text | Aloe