An article to understand the innovation and opportunities of DeFi transactions, lending and derivatives (www.blockcast.cc)

Author: Zheng Jialiang, Research Director of Blockchain Investment Fund HashKey Capital

To the center of the financial (DeFi) industry after more than two years of development, it has been gradually show a strong ecological, although this eco-based Ethernet Square, but not exactly equal to the ecological DeFi Ethernet Square. We thought last year that DeFi will move closer to traditional finance and learn from many standard designs. But relying on the characteristics of decentralization and no censorship , DeFi has developed a different look.

This article summarizes the brief points of the DeFi track as follows:

• DeFi’s AMM is a new way that can make up for the deficiencies of the centralized exchange (CEX) and has great potential for development. In the past, present, and for some time in the future, trading will still be the number one demand in the blockchain world;
• The rising demand for borrowing comes from liquidity mining and nested leverage, which have a function of amplifying revenue;
• Decentralized derivatives are more complicated. No mature platforms have been seen in decentralized exchanges. Derivatives and synthetic products combined with contracts will develop;
• There are still very few participants in decentralized insurance, the industry’s demand is very rigid , and the driving effect of the track is obvious. The return on risk investment will be great.

We also briefly did a quantitative analysis, studied the income characteristics of DeFi tokens, and found several interesting phenomena:

1. According to the current data, the price gains of DeFi tokens have little relationship with short-term mining gains (a bit counter-intuitive, which will be explained later);
2. Surprisingly, the income of DeFi tokens has little to do with ETH ;
3. There are two categories that have the greatest relationship with the income of DeFi tokens. One is Bitcoin , which has obvious negative correlation characteristics. The other is that it has the same direction as the non-DeFi altcoins with small market capitalization, indicating that DeFi has more altcoin characteristics;
4. Our statistics show that there are still about 33% of the changes that cannot be explained, which may be attributed to factors such as valuation, fair launch and long-term mining revenue, and project fundamentals.

We judge the prospects of this liquidity mining as follows:

• Liquid mining is a way to make good use of profit-seeking onboarding users, like the early subsidies of Uber, Didi and Internet products;
• Reflects a certain degree of fairness, allowing the community to make full positive contributions and earn more with more work;
• But it is still inevitable to be controlled by predators , predators are like big miners in a mining pool, and those with strong capital benefit the most;
• The hastily launched community projects are more dangerous. The actual value of project tokens has not been tested, and the economic models of many projects have not been straightened out. Tokens have only governance functions , which are difficult to capture value and are covered by the FOMO atmosphere;
• The later the project is more difficult to go online, the taste of the community is getting higher and higher, and there will be a situation of “picking eyes”. Refer to the later stage of the IEO (this is the case now);
• Certain valuable products and economic models will be left behind, but not the majority;
• Ethereum’s fee congestion is actually a foil to other public chains , and the next-generation public chain competition increases uncertainty.

Transaction: AMM mechanism organic solution

Coin listing, cold start, slippage, liquidity issues

We believe that the most important breakthrough of DeFi is the decentralized exchange .

The idea of ​​DEX in the past was to directly move CEX’s order book model to the chain. Although it can realize the self-managed assets of the blockchain and the absence of KYC/AML, it ignores three problems:

1. This mechanism has no advantage compared with CEX, because the speed of the order book cannot catch up with the centralized order trading center, and there is no way to find the price in time;
2. The current structure of Ethereum has extremely slow transaction speed;
3. Gas fee is also required to create an order (no transaction will not be returned). So basically DEX can only appear as the role of CEX’s copycat.

AMM, liquidity mining and project startup mechanism are closely integrated

However, the advantage of DEX is actually that it can quickly obtain open market liquidity , and if you want to list on CEX, you need to go through a long process and huge overhead (at least before). It can be said that the CEX model determines that it is impossible for the project party to operate in a decentralized small team or community model. A complete set of financing, issuance, currency listing, and market making processes is very difficult to complete without a team to do a cold start.

DEX can correspond to the mode of community currency issuance (now beginning to be standardized slowly). A DeFi project token does not have a strong team to operate, and it is difficult to get through CEX, but it can quickly realize trading pairs on DEX. CEX can’t help it.

In addition, the cold start of the DEX order book model is very complicated, requiring the market making team to understand not only market making, but also blockchain, smart contracts, wallet interaction, etc. But AMM solves this cold start problem. It only needs to provide LP token , and ordinary people can participate in market making, which greatly reduces the threshold. With the addition of the incentive layer (liquid mining), a series of problems (at least partially solved) such as currency listing, cold start, and transaction slippage can be solved (at least partially), and unexpected results have been obtained.

AMM’s price discovery mechanism is still weaker than centralized exchanges

Of course, there are many problems with AMM. Here we put the comparison between common AMM and order book mode:

The biggest problem with AMM is impermanence loss , because AMM does not take the initiative to provide orders, but only to provide a pool. The centralized order book is where the maker and taker provide liquidity based on a certain market price, and spontaneously do what the oracle does. The centralized exchange is essentially an order book sorting and matching pool, and the price is determined by the order. The initiator decides for himself.

AMM, such as the popular constant product market maker CPMM , the price is determined by the number of tokens, that is, the constant product of X*Y determines that ∆X can be exchanged for ∆Y, then after the transaction is exchanged, the corresponding token price is known How much. Therefore, the price estimate provided by AMM before the transaction is calculated based on the ratio of ∆X and ∆Y.

Therefore, the real price discovery of AMM must be seen after the transaction , rather than forming market prices spontaneously through the maker and taker like a centralized order book. So arbitrageurs take advantage of this to make a profit, thus causing impermanence losses. However, some AMMs have begun to introduce the oracle mechanism to provide prices, but the quotation frequency of the oracle mechanism needs to match the transaction frequency. The price cannot be adjusted automatically like an order.

Borrowing: Provides liquidity, leverage and income amplifier

Lending is the earliest DeFi, represented by MakerDao . Compound, Aave, bZx, etc. were launched later. Compound ‘s liquidity mining ignited the entire community and became a pioneer in the DeFi industry.

The minimum amount of borrowing is actually related to trading and liquidity mining:

1. You can mine on your own lending platform, such as Compound, bZx, etc.;
2. The lending platform provides a leverage tool. One type of asset can be exchanged for another type of mineable asset, or continuous borrowing for liquid mining, because the mining revenue is higher and the borrowing cost can be fully covered. The lending platform has become a revenue amplifier;
3. If it is possible to carry out credit loans such as Aave, the increase in leverage will be even greater.

However, there is a prerequisite for the above starting volume, that is, the end point of the entire mining game-the profitability of liquid mining must be relatively stable . The current industry is constantly extending the game forward (life extension). Although I don’t know where it can be extended, including centralized exchanges that have begun to carry the banner, it is already a bit difficult.

But through mobile mining, the product capabilities of the loan agreement are activated: increase the currency, increase the loan function (credit), liquid mining for user subsidies, obtain a group of real users, conduct user education and stay The available products.

There are also some lending protocols that try to introduce a centralized method, that is, using real identity information (such as banks, telecommunications), etc. for credit lending, such as Teller. This can also be seen as the traditional centralized lending to change the product into a digital currency.

Insurance: great potential but rough product

Decentralized insurance is the most innovative mechanism of DeFi. Compared with other types, it is still at an early stage and has a greater opportunity. We discussed the topic of insurance mechanisms in the last report. In traditional insurance, the form of joint-stock insurance companies is the mainstream (50%-70%). Mutual assistance and mutual insurance are secondary forms, but the proportion is not low, accounting for about 27% of the global market. In the United States And Europe is particularly developed, both exceeding 30%, Japan exceeding 40%, and China’s share is very small.

The advantages of mutual insurance companies compared with joint-stock insurance companies are: joint-stock insurance companies involve the interests of three parties , that is, managers, policy holders and shareholders. In a sense, policy holders and shareholders have opposite interests, that is, more compensation For policyholders of 1 yuan, the shareholders’ benefits will be reduced by 1 yuan. However, it is alleviated through the mechanism of checks and balances : that is, if the insurance company is obviously biased towards shareholders, the number of insurance purchases will be greatly reduced, affecting long-term interests.

Mutual insurance only managers and policy holders two parties, policyholders into a pond, all compensation out of the pond, does not involve the shareholders, greatly simplified. Therefore, the nature of mutual insurance and blockchain is similar in some cases. For example, the owners of blockchain projects are only divided into two types: teams and general token holders.

Centralized insurance can effectively use the mutual insurance model

Although the number of interested parties involved is reduced, it is not easy to coordinate the interests, because the risks of different forms of insurance vary greatly. Taking Nexus Mutual as an example, insurance is provided for different contracts, but the situation of unintended use in different contracts is definitely very different. Therefore, between cover holders, they must try to claim more compensation for themselves and less accompany others. (The nature of people seeking advantages and avoiding disadvantages). At present, only bXz is the only example of claim settlement on Nexus Mutual. Therefore, for Nexus Mutual to be promoted on a large scale, the claim mechanism needs to be carefully considered. The current prosperity can hardly be said to be the success of a mechanism or product (after all, it is rough and not very effective, and the loss determination of traditional insurance is more standardized), but it is a success in opportunity.

Nexus Mutual also hopes to push insurance products into the wider real world. For example, the amount of insurance is not linked to actual losses. For example, compensation can be partially processed instead of simply yes/no. All this requires more efforts in the direction of products and pricing, which is a very test. The team has deep insurance and actuarial experience , so there is still great hope for the intensive cultivation of products.

Of course, another idea is to use financial products such as options for certain protection, but this is more of a type of derivatives than insurance.

Derivatives: The synthetic product track has many advantages

Synthetic products have more advantages than traditional finance

We feel that DeFi derivatives can be differentiated from CeFi by synthetic product series. Synthetic products are a special kind of products (of course they will be regarded as derivative products), but synthetic products have been realized in traditional finance. A large number of transactions have been conducted in the open market through CFD (spread platform), and in the non-open market through investment banks. A lot of customized services have been carried out, and they are not completely new.

The risk of traditional CFD platforms is mainly counterparty risk . In fact, the corresponding assets cannot be actually held through the CFD platform. The CFD platform does hedging in the middle, resulting in the actual counterparty between the user and the platform.

If it is done with DeFi, the counterparty risk is actually reduced. This is determined by DeFi and blockchain mechanisms. Take Synthetix as an example. Because all products are based on the native tokens of the user mortgage platform, to make the risk of synthetic assets controllable, over-collateralization (750%) is required. The synthetic asset minter will receive the transaction fees of the asset and must Only by maintaining this mortgage ratio can you claim the cost share, and more synthetic products and transactions are the key to the success of the platform, so the platform’s motivation for evil is reduced. In addition, the current CFD platform still requires a traditional KYC process, and the participation process of the DeFi synthetic asset platform is smoother.

Perpetual contracts have the largest profit margins, and new designs and structures are needed

Another trend in DeFi derivatives is that the market starts to target perpetual futures contracts . Previously, Lianwen summarized the six DEX platforms launched. In fact, those who do perpetual contracts see that in CEX, the development of contract transactions is very fast, and it has surpassed spot transactions. On March 12, the BitMEX server was down. It was also seen that some DeFi could only have contract liquidation problems. The DEX of perpetual contracts has not yet received such a test. So based on the current Ethereum architecture, the performance may not be good enough.

I also see that other high-performance public chain DEXs are also slowly participating. This is very interesting. Products such as Cosmos, Polkadot, and Solana are very powerful competitors. It will also stimulate the cross-chain trading track.

Derivatives are definitely the most profitable piece of CEX. CEX will not relax its pursuit of products, market share, and mechanism design. DEX derivatives are equivalent to encountering the most difficult competitors. But it is also possible to design a “similar” liquidity mining mechanism, which requires more whims and opportunities, otherwise it may move to a similar pattern of DEX with a thin order to compete with the spot exchange CEX.

Analysis of DeFi token income characteristics

We will do a very preliminary analysis here and try to find out what factors come from the fluctuations in the price of DeFi tokens. Here we use the method used by Zayn Khamisa in the article “An analysis of the factors driving performance in the cryptocurrency market”.

Here, as the dependent variable, DeFi tokens do not use any single token to prevent the large fluctuation of a single token from being too large. We choose the DeFi index on the exchange FTX as the dependent variable, which contains 11 DeFi tokens. (DeFi Pulse also launched a similar DeFi index, but with too few data points)

We select several types of independent variables: one is the DeFi mining yield , the second is the BTC price, the third is the ETH price, the fourth is the large altcoin index Alt, the fifth is the medium altcoin index Mid, and the sixth is the small altcoin index Small. (Among them, four, five, and six are also taken from FTX, and there are five constituent currencies in the six that overlap with the DeFi index and have been eliminated).

The independent variable and the dependent variable are all selected daily rate of return (that is, the daily rate of change), which has been smoothly processed, so the regression method of OLS can be used, imitating the approach in Zayn Khamisa’s article. The following are the simulation results:

We found that only BTC and small altcoin indexes have explanatory power (significant P-Value), and BTC is negative, the rate of return model can be simplified as follows:

Our understanding of this rate of return model is as follows:

1. The income of BTC and DeFi is negatively correlated . It can be seen from the phenomenon that when the altcoin is dancing, the BTC market is slightly flat;
2. Changes or rises in BTC will generally cause the so-called DeFi tokens to suck blood . It may also come from the demand for hedging/profitability, that is, the hedging/profitability of DeFi rushes towards BTC;
3. The income of DeFi coins is very similar to that of small altcoins, so it can be classified as a high-risk altcoin at present;
4. The dependent variable setting of this model also has some shortcomings. Because there is no data for the mining yield, we use gradient descent instead. If there is a real yield, it can be more refined. At present, we can’t see any strong connection, and it may also be related to the variety and the diversified yield. Theoretically, the mining yield will be a plane opened according to the date* variety;
5. The explanatory power of the model reached 0.677, which means that 2/3 of the change can be explained, and about 1/3 of the change cannot be explained. This can also be attributed to the Alpha of the DeFi token revenue. There should be roughly four sources of income. The first is the fairness of the online level . The more decentralized projects the community gives, the higher the evaluation, such as the founder and team without initial token distribution or pre-mining. The second is a pure valuation factor . New themes are always favored by the market, and valuations naturally become higher. The third prediction is long-term stable mining income . Logically, the higher the mining income, the greater the purchasing power, and the more valuable the token, similar to a fixed income product. The fourth is the fundamentals of the project, including technical foundation, team, code quality, community activity, long-term operation time, etc.;
6. What is interesting is that DeFi tokens have little relationship with Ether, that is, neither positive nor negative. If we only start from the model, DeFi tokens have little relationship with the Ethereum ecosystem. However, the design of Ethereum does have a certain deviation from the prosperity of the above applications (that is, weak value capture ability), and there are many types of tokens that can be used for liquid mining, not limited to Ether;
7. Since the data only starts on June 19, there are fewer data points, and more data points can be further classified in the future;
8. You can also add more independent variables, such as other external factors. However, due to the niche nature of DeFi tokens, choosing other categories such as VIX, oil, gold, etc. used in the article by Zayn Khamisa, we think it is of little significance. If the other independent variables are clear, you can further strip the factors from Alpha.

Summary and outlook

This article sorts out and judges the DeFi market to a certain extent. Generally speaking, several sub-categories of the DeFi market have long-term potential. We are optimistic about the directions of automated market makers, synthetic products, and decentralized insurance . Liquid mining has heated up the industry for a while, but it will also cause some problems. In addition to the token innovation mechanism, there will be some valuable business models in the future. The income characteristics of DeFi tokens are similar to other small altcoins, indicating that they are still the characteristics of early blockchain projects. The alpha income part is expected to be related to the degree of decentralization, valuation and long-term mining income and project fundamentals. There is a big relationship.