Dialogue with Wang Wei: Reinterpreting DeFi from the Perspective of First Principles

Recently, the 17th session of the Catcher Academy, hosted by Chain Catcher, invited Wang Wei, co-founder/CTO of Unizon, to give a thematic presentation on "2020 Annual Summary - DeFi Special."

Unizon is a blockchain technology company focused on open finance (DeFi) and digital assets, co-founded by leading figures in blockchain technology and digital asset research and consulting, including Meng Yan, Wang Wei, and Zhou Zhiqiang. The company is dedicated to emerging digital asset research and DeFi protocol development, aiming to create innovative financial infrastructure through blockchain technology, providing convenient and transparent financial service applications to the market.

In the Catcher Academy community, Wang Wei reanalyzed the development logic of the DeFi market in 2020 from the perspective of first principles, sharing in detail the fundamental reasons for this year's market explosion, industry trend judgments, and many valuable insights. He also addressed why there are so few high-quality DeFi applications in China. The full text is organized below, hoping to inspire readers.

Organizer | Henry

1. Review and Reflection

Chain Catcher: In the past year, the DeFi ecosystem has experienced explosive growth, and there has been a lot of discussion around DeFi. In your opinion, what is the biggest misconception people have about DeFi? How should we understand DeFi from the perspective of first principles?

Wang Wei: I think the perspective of first principles is very good. If we don't look at problems from a first principles perspective, it's easy to fall into some misconceptions. For example, there are two typical extreme views on DeFi, both of which belong to this misconception.

One is the question of whether DeFi has real value. Some people believe that the current boom is entirely driven by liquidity mining, and once this enthusiasm passes, DeFi will also fade away. This completely ignores the fact that mainstream projects like Uniswap and Compound had already achieved a TVL of over a billion dollars and trading volumes exceeding ten billion dollars before liquidity mining was initiated.

Conversely, some people believe that DeFi can do anything. For example, a lending platform recently proposed to offer "credit loans," which are loans without collateral or with loan amounts exceeding the value of the collateral. The general counsel of Compound has previously expressed views on this issue, and I have also elaborated on it in a recent article: Credit loans are essentially financial activities based on assessing the repayment ability of individuals in the real world, which is not very suitable for DeFi.

Whether there is value, what can be done and what cannot be done—these core issues can only be reasonably answered by thinking from the perspective of first principles. My view is that the entire financial system can be divided into the "credit" part and the "computational" part, where the credit part is responsible for the expansion and contraction of credit, and the computational part is responsible for the flow and allocation of resources.

The first principle of DeFi is the answer to the question of "why we use DeFi," which lies in the fact that the computational part of the financial system has a better implementation method—blockchain + smart contracts, which is also the value network and its exchange rules. I believe that in the future, the computational part of the entire financial world can and should be realized using blockchain + smart contracts, which will make the entire financial system of society more transparent and efficient, benefiting the public and facilitating effective regulation, making it a relatively ideal model. This is my thinking on the first principles of DeFi.

This involves some in-depth thinking about the financial system. From the development history of DeFi, these differences are reflected: the earliest mature DEX, which is a trading platform, does not involve the credit issue of traders because it is a transaction between buyers and sellers. One party has assets, and the other has funds; if the price is right, the transaction can be completed, so it is basically a computational process that can be directly realized through smart contracts. Secondly, the currently mainstream DeFi lending platforms do not involve repayment credit issues as long as there is excess collateral; the entire process can be realized through smart contracts without human participation.

As for the aforementioned "credit loans," since they involve credit evaluation, they cannot completely rely on DeFi platforms. However, we can further analyze and break down the computational part and combine it with the credit system, which could be a good idea. Looking at areas such as insurance and derivatives, I personally feel that the latest DeFi projects are also developing in this direction—when people effectively separate the "credit part" and "computational part" in a financial scenario through research and analysis, DeFi platforms in these areas will emerge in large numbers, which is a typical innovation process.

Chain Catcher: Many people attribute this year's DeFi explosion to liquidity mining. What do you think is the fundamental reason for the explosion of the DeFi ecosystem this year? From what angles should we view the rationality of various projects' liquidity mining mechanisms and economic models?

Wang Wei: I believe that the main reason for this year's DeFi explosion is the maturity and usability of some mainstream DeFi platforms, as well as the combinability of multiple DeFi platforms in a "Lego" model. Liquidity mining is indeed a very effective catalyst.

For example, Uniswap became popular this year mainly due to the launch of version 2, which introduced customizable trading pairs and quickly increased market activity. In fact, Uniswap had a TVL of several hundred million dollars during its v1 period, and v2 stabilized it at the billion-dollar level (there was some turbulence with Sushiswap in between, but it ultimately returned to this level).

After Uniswap launched UNI mining, the TVL peaked at $3.3 billion, and after it ended, it returned to the $1.5 billion level. From this context, it can be seen that liquidity mining indeed plays a significant role in driving growth, but it does not change the fundamentals and scale of a platform's product. Other platforms like Compound follow a similar pattern.

Each major project's mining mechanism has its own merits; I am not an expert in this area and cannot comment too much. However, I have always believed that liquidity mining mechanisms are an important means during a platform's startup phase or a certain upward breakthrough phase, rather than the core factor for the platform's development and success.

Chain Catcher: This year's DeFi wave has made it very clear that good projects are mostly abroad, leading us to ponder a question: why are there so few high-quality DeFi applications in China?

Wang Wei: In fact, there are also many good projects in China, and there are many excellent DeFi developers and teams. However, since this wave of DeFi originated abroad, China is in the role of a follower/competitor. From this perspective, being behind abroad is an objective fact. DeFi belongs to blockchain, which is inherently open and global; the national boundaries are not as strong as centralized systems. Some DeFi projects being developed in China are also well-received by foreign communities, and many Chinese people are part of the teams behind some foreign DeFi projects.

I think everyone may have their own answers to this question, such as technical level, social environment, regulatory conditions, etc. What I want to discuss mainly is the aspect of the geek spirit. Over the past decade, domestic innovation has mainly been driven by investment, closely related to the rapid economic development of the country. In contrast, innovation abroad has always had a mainline of geek spirit, running parallel to investment-driven innovation, and often taking the leading position.

In the internet field, it seems that domestic innovation is currently more successful, with internet innovations generally moving towards a broader integration with life, such as the recent buzz around tech giants selling groceries. At this level, investment-driven innovation clearly has an advantage. However, investment-driven innovation is not a decisive factor for all innovations, especially for many very early-stage projects, where creativity and realization do not require large amounts of funding; it relies more on innovative ideas or even passion, which is particularly prominent in the purely digital field of DeFi.

To give a somewhat inappropriate example, J.K. Rowling wrote Harry Potter in a café while on unemployment benefits; giving her more money probably wouldn't have changed that. Therefore, both early blockchain and the current DeFi have shadows of cyberpunk behind them, emphasizing "High tech, low life," focusing not on making big money, but on a freer and fairer environment. This aspect may not be easily supported by investors whose primary goal is financial returns.

On a deeper level, this reflects a microcosm of our entire society. The slogan "High tech, low life" is unlikely to appear among people who genuinely live in low life due to lack of money; otherwise, many might ridicule him for having an "A Q spirit."

What I want to say is that the pressure most of our young people face in their daily lives indeed harms their innovative drive. In this larger environment, it is natural to hope for more support from investors.

Chain Catcher: This year, there have been many "innovations" in the DeFi industry, such as the recent algorithmic stablecoins, which have attracted much attention and controversy. How do you view the value of algorithmic stablecoins? How do you distinguish between true product innovation and hype?

Wang Wei: Using this example, I would like to apply the model of dividing the financial system into "credit" and "algorithm" for some analysis. Simply put, algorithmic stablecoins incorporate the concept of "stability" into an algorithmic framework. What is the source of a stablecoin's "stability"? In simple terms, stability is a subjective feeling that comes from our daily lives, where currency serves as the unit of account for all other goods. For example, living in Japan, the yen is a "stablecoin," but in reality, the yen may fluctuate significantly against the dollar.

There are many types of algorithmic stablecoins that I know of, with different design logics, and some are quite sophisticated, so they cannot be generalized. To give a simple example, stock exchanges around the world have the ability to "split" and "consolidate" stocks. Suppose a stock splits every time its price rises to a certain level, returning to the price before the increase; conversely, it consolidates when the price falls to a certain level, returning to the price before the decrease.

In this way, the "nominal price" of the stock can remain unchanged, but if it is a good company that pays out dividends, and everyone buys it, then three years later, while the stock price appears unchanged, investors have actually made a lot of money. Conversely, if it is a bad company with a continuously declining stock price, the conclusion will still be a loss. It can be seen that maintaining a constant nominal price through some algorithm does not make this asset a "stablecoin." This type of algorithmic stablecoin actually forms a game structure that exploits people's hopes of making money while fearing losses.

Therefore, my view is that the concept of "stability" in stablecoins comes from why they are accepted as the unit of account for other assets; this is a concept derived from the external environment and people's subjective thoughts, which actually belongs to the "credit" part of the financial system, rather than the algorithmic part. For example, a country's fiat currency is determined by the national environment. In our digital currency field, USDX or XUSD is determined by many factors, such as exchanges, and algorithms are difficult to play a similar role.

This leads to the second question: how to distinguish between financial innovation and hype? My suggestion is to analyze more from the perspective of financial logic, that is, from the angles of "credit" and "computation," including making more comparisons.

2. Judgments on Trends

Chain Catcher: Multicoin Capital recently pointed out in their research that DeFi projects face various risks, including contract errors, poor protocol parameterization, on-chain congestion, oracle errors, and failures of admin bots/LPs. With frequent security incidents in the industry, to what extent do you think these risks can be mitigated in the future? Will the frequency of DeFi security incidents significantly decrease in the coming years?

Wang Wei: I am a technical person myself. Security incidents can generally be divided into two categories: one is inherent to blockchain, such as on-chain congestion; the other is typical human error, which we usually refer to as program bugs or lack of optimization.

First, regarding the inherent issues of blockchain, the most talked-about problem is congestion. Essentially, if we enjoy the capabilities provided by this special computational system of blockchain, we should pay the corresponding costs or prices, including speed, congestion, and GAS fees. Hoping that Ethereum 2.0 will solve these problems will not yield a direct result.

Let’s make a simple deduction: if Ethereum 2.0 has fast transaction speeds and low fees, then more transactions will occur. From the essence of the blockchain consensus mechanism, the overall average speed will decrease (because a block has limited capacity). At the same time, if the speed of individual transactions is indeed much improved compared to 1.0, people will be more inclined to pay higher fees for speed, while those paying lower GAS fees will not be able to queue and will still encounter congestion. If everyone is competing for resources, GAS fees will rise again. Therefore, ultimately, everything still comes down to market and user choices, not purely determined by technology.

Now, regarding the second type of problem, which is human error, programs are ultimately written by people, and people make mistakes. Many platforms that have been attacked mainly have issues with their design thinking or code logic. I have worked in traditional IT for many years, and I can responsibly say that smart contracts do not have more problems in this regard than traditional IT.

The issue is that the entire blockchain and smart contract system is exposed to everyone, without the protection of traditional firewalls or layered systems, so any problem can be attacked. Therefore, the likelihood of these errors causing problems is higher than in traditional closed systems, and the resulting losses may also be greater. The DeFi industry has been continuously improving in this regard, and it should be said that so far, the effects have been significant, and the entire ecological environment has developed.

Chain Catcher: Now the CeFi ecosystem is also developing rapidly. What do you think will be the respective positions of CeFi and DeFi in the industry in the future? Or what will their relationship look like in the future?

Wang Wei: On August 30 of this year, Uniswap's trading volume first surpassed Coinbase, and in September, its total trading volume also exceeded Coinbase. I believe that in terms of DeFi being able to achieve products with capabilities equivalent to CeFi, the overall trend is still developing towards surpassing or even replacing CeFi.

The rapid development of CeFi is generally driven in two directions: one is that some product logics are relatively complex, and at this stage, the efficiency of implementing them through smart contracts is not high, making it difficult to achieve the desired effect; the other is that centralized systems have the ability to act as counterparties, which is the credit capability we have always emphasized. For example, in certain derivatives, the platform may not be a neutral party but has its own funds to operate. The former is mainly a technical issue, and I think this is where Ethereum 2.0 may exert its efforts; otherwise, other public chains may catch up. In any case, if technical issues can be resolved, it will not change the overall trend of DeFi catching up. The latter is what we mentioned earlier: there are areas that DeFi cannot cover because relying solely on algorithms cannot solve credit issues.

Therefore, I believe that CeFi and DeFi should form a relationship of both competition and cooperation in the future. The competitive points will mainly be in functions like trading and collateralized lending that both can achieve, where DeFi's advantages will definitely grow. The cooperation points will be in different links of the long chain of asset issuance, circulation, and clearing, where everyone can play their roles and complement each other. In this, I want to emphasize that CeFi's main direction must be to connect with the real world, including fiat currency digitization, credit lending, and physical asset collateralization, which must rely on information or people's credit in the real world to complete.

I believe that the ultimate trend is that there will no longer be a distinction between CeFi and DeFi. For example, Maker is introducing real-world assets as collateral, which inevitably involves a centralized management process. Can we say that Maker is also entering CeFi's territory or has become CeFi? This brings us back to the initial question about the first principles of DeFi: in the future, there will no longer be a distinction between CeFi and DeFi, but rather a distinction between "credit" and "computation" within the financial system.

Chain Catcher: Mergers and acquisitions among leading DeFi projects represented by YFI are becoming more frequent. What impact do you think this will have on the industry?

Wang Wei: DeFi platforms are essentially a set of smart contracts, code running on the blockchain. Is project acquisition just code acquiring code? One smart contract acquiring another smart contract? Clearly not possible, so what is actually being acquired is still people and teams.

In the past, acquisition targets might have included intangible assets such as code, intellectual property, patents, etc., but in the open and transparent world of DeFi, even these are absent. If we must say there are still intangible assets, they would be the ideas in developers' minds that have yet to be realized, as well as the code release permissions of existing operational platforms, including GitHub repositories, platform management addresses, private keys, etc. From this perspective, the purpose of acquisitions is quite clear: it is for team integration/collaboration.

Further analysis shows that DeFi protocols are inherently connectable and stackable. What benefits can team integration bring? There is only one benefit: through teamwork, the connections between the acquired and acquiring projects can be optimized, whether it is simplifying interfaces and processes, reducing gas fees, or coordinating user benefits between protocols.

Assuming that the acquisitions indeed produce such effects and that such acquisitions become more frequent, the result is likely to transform the DeFi world from a collection of independent protocols into a "protocol family" with strong internal interconnections and high collaborative efficiency.

Once this trend occurs, it will have profound implications for the development of the DeFi industry, whether it is a good thing or a bad thing remains to be seen, requiring continuous observation and in-depth thought.

Chain Catcher: For the DeFi industry in 2021, what do you think are the three most important trends?

Wang Wei: I believe there are three trends or directions for the DeFi industry next year:

The first trend is "innovation," as the DeFi industry is still in its early stages, with many new areas to explore. The innovation referred to here is about continuously breaking down the "computational" part of the financial field mentioned in the first principles of DeFi. Whether it is insurance, interest rates, or derivatives, there are many directions for innovation, but the direction of innovation must be grasped. This is the first principle of DeFi development; mastering this rule will lead to effective innovation and success; going against it, regardless of any new ideas, will at most be fleeting.

The second trend is "security." 2020 was a year of DeFi explosion, but it was also the year when DeFi faced systemic and large-scale attacks. From the beginning to the end of the year, attack incidents occurred continuously, resulting in losses of hundreds of millions of dollars. However, I believe that new things must go through this process. As the rules of smart contract security are increasingly mastered, auditing capabilities become stronger, and pricing mechanisms become more rigorous and reasonable, the security of DeFi in 2021 will significantly improve, which is crucial for the healthy development of DeFi.

The third trend is "integration." Integration has many layers, such as the project mergers mentioned earlier, which is also a form of integration. But what I want to emphasize is the integration with real assets. This year, Maker began to introduce off-chain assets as collateral, and many projects are exploring how to combine real economic financing activities with DeFi.

Originally, I felt that it might take another 2-3 years for this situation to emerge, as DeFi is suitable for native digital assets. Introducing real assets too early could lead to blindly repeating the mistakes of the past few years with "blockchain+" initiatives. However, considering the rapid development of DeFi and the fact that native digital asset DeFi has already entered its third year, next year may mark the beginning of a new phase, where this new phase is the integration with off-chain assets.