Partnered With:

Zou Chuanwei: Re-understand DeFi from eight key perspectives and logic

Read this article in 39 Minutes

In what direction will the DeFi ecosystem evolve? What logic does the evolution of DeFi follow?

Original title: "Zou Chuanwei: Re-understanding DeFi from eight key perspectives and logic"

Original source: Lianwen

Original author: Zou Chuanwei, Chief of Wanxiang Blockchain Economists

There have been a lot of discussions on DeFi recently, and DeFi can be understood from multiple perspectives. This article proposes to understand DeFi from 8 key perspectives : 1. Financial functions; 2. Discrete time finance; 3. De-trusted environment; 4. Oracles and information; Incentives; 8. Risks and interrelationships.

This article attempts to answer two questions:

1. DeFi developers, in addition to writing smart Besides contracts, what other tools should you master ?

2. From the perspective of DeFi investors, in what direction will the DeFi ecosystem evolve, and what logic will the DeFi evolution follow?

Financial functions

Some researchers refer to mainstream financial institutions to understand DeFi, such as discussing the DeFi field What should be the form of banks, securities and insurance, etc. This analogy is actually not accurate enough, because DeFi is built according to financial functional modules . This involves the relationship between financial institutions and financial functions.

Financial development has two main lines intertwined and ascending like DNA double strands, one is financial function and the other is financial institution. Zvi Bodie and Robert Merton have proposed 6 basic financial functions:

* Payment and settlement . This is the most basic function of financial services for real economic activities. In addition to barter, any economic activity depends on payment and settlement to form a value closed loop.

* Aggregation of resources and equity segmentation . In this way, social resources can be gathered to do larger-scale or more adventurous undertakings, and the benefits and risks can be shared among participants. A joint-stock company is a typical embodiment of this function. Banks gather scattered funds from society to support national development and construction, which also reflects this function.

* Transfer resources across time and space . Resources are never evenly distributed in time and space, and resources need to be adjusted in two dimensions to maximize resource usage efficiency . For example, Zhang San has spare money and cannot spend it, and Li Si has a project but is short of money. Through the financial system, Zhang San can lend money to Li Si (cross-space transfer) . In the future, after Li Si's project obtains income, part of the income will be given to Zhang San (transferred across time).

* Manage risk . The allocation of resources in the financial system is carried out in an uncertain environment. The essence of any financial activity, whether it is centralized or decentralized, is to obtain benefits through operating risks.

* Provide information. The financial system has an important price discovery function for money and risk. The information provided by the financial system is of great significance to the resource allocation of the whole society.

* Solving Motivational Problems . A large amount of value in the whole society is traded and circulated through the financial system, and the resulting economic incentives promote the formation of social division of labor and market order.

Mainstream financial institutions generally perform multiple financial functions at the same time. For example, banks mainly perform functions such as payment and settlement, pooling resources, transferring resources across time and space, and managing risks.

Bodie and Merton argue that financial functions are more stable than financial institutions and that financial institutions take form depending on the financial functions they perform. For example, the organizational structure and business form of banks 100 years ago are very different from today's, but the financial functions performed by banks have not changed much.

The division of financial functions by Bodie and Merton is also applicable to DeFi (Table 1) :

< /p>

Zou Chuanwei: Re-understand DeFi from eight key perspectives and logic

Therefore, to understand DeFi, it is best to start with financial functions . DeFi is built according to financial functional modules and has good composability. The assembly of multiple DeFi projects can achieve complex and diverse financial functions, which can be close to mainstream banks, securities and insurance. But even the DeFi portfolio is very different from these mainstream financial institutions.

Discrete-time finance

Mainstream finance is continuous-time finance. For example, global foreign exchange transactions occur 24*7, and Yu'e Bao generates interest income every day. The time units used by mainstream finance are generally hours, days, weeks, months, quarters, and years.

DeFi is discrete time finance . Any financial activity has cycle and frequency. The frequency of DeFi depends on the update frequency of the distributed ledger of the public chain. The extension of time in the public chain is reflected in the continuous generation of blocks, and the unit of time is the block time. The block generation time of the public chain is determined by the consensus algorithm run by the verification nodes in the distributed network. Although there is a statistical average, it is a random variable in the future.

* Discrete time and TPS constraints have a comprehensive and profound impact on DeFi:

* Affecting the efficiency of DeFi activities, the amount of DeFi activities is naturally subject to the physical performance of the public chain . For example, when the chain is congested, transactions such as on-chain auctions and on-chain collateral disposal may not be processed in a timely manner.

* Affects the efficiency of intra-chain and off-chain information synchronization and arbitrage. This will be elaborated later in the discussion of oracles and arbitrage.

* Affects the efficiency of price discovery and risk clearing . For example, except for long-tail encrypted assets, the price discovery of most encrypted assets is done on centralized exchanges. DEXs follow pricing from centralized exchanges, not the other way around. For another example, when the market panics, the public chain is prone to congestion , and some transactions that help to clear market risks may not be processed by miners, or require higher handling fees or gas fees to be processed by miners . This will not only reduce the efficiency of market risk clearing and rebalancing, but also reduce market participants' confidence in the orderly operation of the market and further amplify market panic .

What analysis method should discrete-time finance adopt? Continuous-time finance is easier to analyze than discrete-time finance because of the use of mathematical tools such as calculus . This paper proposes the following analysis method for DeFi:

_ Discrete time finance = continuous time finance approximation + public chain TPS and time lag influence_

In other words, when analyzing DeFi, the core financial issues should be extracted first, and after a clear analysis in continuous time, the impact of public chain TPS and time lag should be considered. For example, the analysis of the arbitrage mechanism in automatic market makers and oracles can follow this methodology.

In discrete-time finance, the time value of money still applies. For example, after 1 year of Staking one unit of encrypted assets, you will get a total of 1.5 units with principal and interest. Then, 1 unit of encrypted assets today is equivalent to 1.5 units of encrypted assets in 1 year. This is the time value of money . The analysis of capital cost and investment income in DeFi is essentially analyzing the time value of money.

Although the block time is a random variable in the future, taking the block time as the time unit can introduce the interest theory of mainstream finance into DeFi. Basic concepts and tools such as present value, future value, discount factor, simple interest, compound interest, and arbitrage-free pricing are applicable to DeFi and have shown strong application value in many problems

Yebao can query investment returns in real time, but DeFi cannot do this due to the performance limitations of the public chain. In many DeFi applications, theoretically, each new block is accompanied by the generation of interest. However, if these newly generated interests are paid through on-chain transactions at the first time, it may consume high gas fees and cause on-chain congestion. Therefore, it is often necessary to appropriately extend the interest payment cycle . For example, first accumulate the interest instead of immediately transferring it to the relevant address through on-chain transactions, and then pay the interest income in one go after a period of time (such as when exiting staking) . In such problems, it is more necessary to introduce accurate calculation methods of interest.

Interest theory has important applications in Staking and DeFi liquidity management. For example, PoS mining pools also need provide liquidity while providing investors with Staking income. Especially in Ethereum 2.0, the staking lock-up period on the beacon chain is very long. When investors participate in beacon chain staking, if they have liquidity needs, can they transfer their staking shares? This problem is a good application of the theory of interest.

Trustless environment

No matter what role DeFi participants take on, they are essentially It is the address in the public chain . The public chain is a trustless environment, and the address is essentially anonymous , with neither identity nor reputation. This is the key difference between DeFi and mainstream finance. Participants in mainstream finance are individuals and institutions. When an individual or a company applies for a loan from a bank, the bank will assess their repayment willingness and ability; earnings outlook. The above evaluation work happens all the time in mainstream finance, but it does not exist in DeFi.

A trustless environment is the foundation of DeFi's openness and permissionlessness. But in a trustless environment , because the address itself cannot become a credit subject, the performance of financial contracts depends on over-collateralization and Staking. For example, suppose address a wants to transfer 1 unit of encrypted assets to address b in the future. To ensure the enforceability of this commitment, address a needs to provide more than 1 unit of encrypted assets as collateral. How to understand overcollateralization and Staking?

* Overcollateralization and Staking are important channels for public chains to capture value from DeFi. Without this mechanism, the price interaction between the public chain and DeFi may be disconnected.

* Overcollateralization locks liquidity, which is equivalent to converting the credit risk of the address into the liquidity risk of the collateral. In both DeFi and mainstream finance, risks will not disappear out of thin air, and in many cases it is just a form of transformation.

* Because of overcollateralization, the risk pricing efficiency of DeFi lending is very low . This is reflected in the fact that the DeFi lending rate does not include a risk premium for the borrower and has nothing to do with the creditworthiness of the borrower. In fact, in a trustless environment, there is no way to define or measure the credit of an address. It can also be said that DeFi lending is essentially a mortgage loan, not a credit loan.

* In MakerDAO , over-collateralization guarantees Dai liabilities) have the same value connotation. Regardless of who initiates the CDP, and regardless of the collateral used by the CDP, as long as the over-collateralization rule is followed, Dai is equivalent to each other.

* Staking is a stakeholder commitment mechanism . This has been fully reflected in PoS-type consensus algorithms to solve the Nothing at stake problem.

Financial activities cannot do without trust. Trust can reduce uncertainty about the future and is crucial to reducing transaction costs in financial activities. This holds true for both mainstream finance and DeFi. We say that the blockchain is trustless, and the essence is to transform trust in people and institutions into trust in algorithms and smart contracts , but it is still trust in essence. Introducing trust in a trustless environment can help further reduce DeFi transaction costs. There are three ways introduce trust in DeFi.

*Addresses are associated with off-chain identities and reputations, such as Gitcoin Grants using GitHub accounts against multi-identity attacks and collusion attacks. The controllable anonymity of the central bank's digital currency is essentially to associate the original anonymous address with the real identity through KYC.

* Repeated game in the public chain can suppress opportunistic behavior and form a reputation within the chain, such as ChainLink nodes. The repeated game makes the long-term loss of opportunistic behavior exceed the short-term gain, so that anonymous addresses can also abide by the rules evolved from the game.

* "The Invisible Hand" - profit maximization behavior driven by arbitrage and economic incentives . This is the basic logic of mechanism design. We don’t need to know the values of DeFi participants, such as whether they keep their promises, as long as they are rational economic people, we can speculate on their behavioral characteristics through economic analysis. Conversely, through arbitrage and economic incentive design, DeFi participants can exhibit the behavioral characteristics we expect. This will be discussed in detail later.

Oracle and information

There are two consensus mechanisms inside and outside the public chain. The first one is consensus algorithms such as PoW and PoS , which is to form a consensus on the information originally in the chain within the chain. The second is Oracle Machine , which is to form a consensus on information outside the chain within the chain. The oracle machine is the basis for information synchronization and arbitrage inside and outside the chain. No matter what kind of consensus mechanism it is, it means entropy reduction (that is, eliminating chaos) and requires energy input (or cost consumption) . The goal of oracle design is to minimize the ratio of error to cost.

There are many designs of oracles, but they can be roughly divided into two categories.

The first type of oracle is based on reputation and voting , represented by ChainLink . This type of oracle relies on multiple quoters, and selects the average or median of multiple quotes as the oracle quote to control the error of individual quotes. This type of oracle machine also eliminates irresponsible and malicious bidders through reputation mechanisms and repeated games.

The second type of oracle machine is based on trading and arbitrage , through the arbitrage mechanism, the oracle machine quotation converges to the market price. The use of such oracles for arbitrage mechanisms will be discussed in detail later.

From the perspective of communication engineering, no matter what form it takes, the DeFi oracle is essentially a sampling process with errors and time delays. The oracle samples the continuous signal outside the chain at discrete time points, and then reads the discrete signal into the public chain (Figure 1) .

Zou Chuanwei: Re-understand DeFi from eight key perspectives and logic

Figure 1: The oracle as a sampling process

The oracle relative to the original continuous signal The error for is made up of two parts (Figure 2) . First, the error of signal source . The essence of various oracle solutions is to reduce this error as much as possible. Second, the fluctuations in the original signal , but amplified by the sampling interval and the time it takes to reach consensus. Both of these parameters are largely affected by the performance of the public chain.

Zou Chuanwei: Re-understand DeFi from eight key perspectives and logic

Figure 2: Oracle error decomposition

Liquidity

Liquidity is the core issue in DeFi. The application and management of liquidity is reflected in many aspects of DeFi. For example, as discussed above, over-collateralization converts credit risk of addresses into liquidity risk of collateral.

Liquidity reflects the likelihood of selling an asset at a reasonable price within a reasonable time. All else being equal, the longer the time frame, the more likely the asset will be sold at a reasonable price. But in many cases, it is not possible to sell assets in a leisurely manner. In this way, the level of liquidity will have a great impact on the interests of investors.

Mobility is a complex economic phenomenon affected by many factors. For example, the liquidity of intra-chain transactions is limited by the TPS of the public chain. Liquidity is more a product of the interaction between buyers and sellers. The higher the confidence of both parties, the greater the liquidity.

Liquidity is a special public good . For most commodities, higher demand stimulates supply by pushing up prices. However, because liquidity is related to the confidence of buyers and sellers, it may be least supplied when it is most needed.

For investors, liquidity is a promise and insurance mechanism. Liquidity provides investors with confidence that a deal will be closed and at what price.

There are two main types of commitment mechanisms related to liquidity. The first category is reputation-based commitments, such as market makers under an order book. Based on their own reputation and strength, such market makers promise to provide transaction convenience for buyers and sellers and make profits through the bid-ask spread. The second category is algorithm-based commitments, such as automated market makers . Automatic market makers provide buyers and sellers with the convenience of algorithmic transactions through liquidity pools, but they will bear impermanent losses caused by arbitrage and face challenges in business sustainability.

Obviously, under the same conditions, products that can provide liquidity are more attractive to DeFi investors. For example, the PoS mining pool discussed above provides an example of liquidity shares and transfer mechanisms for investors.

Mobility can be unbounded or bounded. Liquidity on centralized exchanges is unbounded. The liquidity of automated market makers is bounded. For example, no matter how investors trade with Uniswap , if the impact of transaction fees is not considered, the liquidity pool should always meet the constant product condition .

PoS mining pool Providing liquidity shares for investors is also bounded liquidity. Regardless of how investors transfer liquidity shares, the total amount of liquidity shares they hold together remains unchanged-the transfer of liquidity shares does not constitute a redemption of the PoS mining pool. This is similar to secondary market trading of shares without affecting the number of outstanding shares of a listed company. As Keynes pointed out, there is no liquidity of investment for society as a whole.

Mobility has agglomeration effect . For multiple liquidity pools, their combined liquidity will exceed the sum of their respective liquidity. This, like the risk diversification effect, is a fundamental law of finance—the risk of a portfolio is less than the sum of the risks of its parts.

Arbitrage

Arbitrage is not a bad thing. Arbitrage is derived from the basic needs of human nature. When conditions permit, everyone wants to take advantage of others, but he does not want others to take advantage of himself. This is human nature.

There are multiple drivers of financial development, such as regulation and technology, but the fundamental driver is arbitrage. When any financial market or product is first launched, because the pricing mechanism is not perfect , there will always be arbitrage opportunities, which will attract arbitrageurs. Driven by arbitrageurs, the pricing mechanism is corrected, and financial markets and products tend to be perfected. Such a cycle goes on and on, and financial development can continue to move forward.

Arbitrage makes price converge, but convergence takes time and cost. For example, for any oracle based on transactions and arbitrage (such as Uniswap) , by solving the optimization problem, it can be proved that there is a no-arbitrage condition under which no new transactions will occur. And this no-arbitrage condition can be equivalent to limiting the range in which the oracle quotes deviate from the market price.

The economic intuition here is very easy to understand. Once the oracle quote deviates from the market price, it means an arbitrage opportunity, but the arbitrageur needs to pay a cost to execute the arbitrage strategy. Therefore, arbitrageurs will carefully weigh the arbitrage benefits and costs, and only execute the arbitrage strategy when the oracle quote deviates from the market price by a large enough margin (this is essentially a question of the optimal exercise time of American options) . The execution of the arbitrage strategy will correct the deviation of the oracle quotes from the market price until the arbitrage strategy is no longer economically attractive, and so on.

It can also be proved that the lower the transaction cost of DeFi , the higher the efficiency of arbitrage source of error) .

Arbitrage-free pricing + interest theory is the basic tool for DeFi asset pricing. Arbitrage will form the interest rate benchmark curve in DeFi. For example, for PoS-type encrypted assets, the staking rate of return will constitute the "anchor" of the DeFi lending rate.

Arbitrage is a universally applicable mechanism design. Arbitrage makes the fewest assumptions about human nature . Arbitrage only requires a person to be rational, to seek advantages and avoid disadvantages, and to maximize his own interests, without knowing whether he is a good person or a bad person. In the decentralized environment of DeFi, this incentive and coordination role of arbitrage is even more important. For example, MakerDAO collateral liquidation is based on arbitrage design.

The prerequisite for the arbitrage mechanism to work is an active community of arbitrageurs , so community incentives are very important. For example, for an oracle machine based on trading and arbitrage, if there is only one arbitrageur, he will wait until the price of the oracle machine deviates greatly from the market price before executing the arbitrage strategy. If there are multiple arbitrageurs, each arbitrageur will consider the possibility that other arbitrageurs will execute the arbitrage strategy ahead of them. Competition among arbitrageurs will advance the execution time of arbitrage strategies, thereby reducing the deviation of oracle quotes from market prices.

No matter what form the arbitrage mechanism takes, arbitrage is essentially a zero-sum game , which is the redistribution of interests, and what A gets is what B gets lose. For example, in an automated market maker, the profit of the arbitrageur corresponds to the impermanent loss of the liquidity provider.

Some automatic market makers introduce oracle machine quotations , which essentially limits the arbitrage space and reduces the impermanent loss of liquidity providers. Obviously, in any arbitrage mechanism, if one party continues to lose money, the arbitrage game cannot continue, and there will always be a day when it stops.

Incentives

Incentive mechanism design should make DeFi an infinite game , and Non-finite game. Community self-organization and self-upgrade are the key to the evolution of DeFi. Community members should be able to obtain their own benefits from DeFi. In other words, in the design of the DeFi incentive mechanism, one cannot expect a certain type of participant to always act as a "living Lei Feng".

As mentioned above, in the automatic market maker, the profit of the arbitrageur corresponds to the impermanent loss of the liquidity provider. In order to make up for the losses of liquidity providers, automatic market makers charge transaction fees from investors, and transfer the fee income to liquidity providers. However, due to the limited transaction volume in the chain and the low standard of handling fees, the " handling fee income < impermanent loss" of liquidity providers is a common problem, and liquidity providers are equivalent to providing liquidity as a public product at their own expense.

Some automatic market makers introduce governance tokens Persistent problem. But the value capture ability of governance tokens is weak. For example, in the valuation of a company's stock, it is generally estimated that the company future profit and dividends , and then estimate the future cash flow of stock holders, and then discount to the current stock valuation. If the investor's equity ratio does not reach critical points such as 33% or 50% , the value of the voting rights corresponding to the stock will generally not be considered. Therefore, the effectiveness of governance tokens for automated market makers remains to be seen. The future direction should improve the charging method and governance token design in the automatic market maker.

The above problems are common in the blockchain field. For example, if the block reward is not considered, or the block reward is reduced to a very low level, can the transaction fees obtained by PoW miners make up for the mining cost? For another example, can the fee paid by the user when calling the oracle machine quotation make up for the cost of the oracle machine? These issues are essentially related to provision and financing of public products in a decentralized environment . To solve these problems, we should refer to the economic theory of public goods.

Risk and interconnection

The core of DeFi is business risk , the main Including market risk, liquidity risk, technology risk and credit risk, etc. Market risk comes from fluctuations in the price of encrypted assets. In DeFi, the common application of over-collateralization and Staking converts the credit risk of addresses into the liquidity risk of collateral, so credit risk is not as prominent as in mainstream finance (bank and corporate bond markets are mainly credit risk) ;. The technical risk of DeFi is much more prominent than that of mainstream finance, which may come from the loopholes of smart contracts or the limitation of public chain TPS.

All kinds of DeFi activities are essentially to maximize returns by taking risks . Risk can be transferred, shared, hedged, switched and spread out, but it never disappears.

The DeFi project is built for financial functional modules and has assemblability . DeFi projects are connected and combined through channels such as information, funds, and risks. This helps to develop the DeFi ecosystem " from point to area ", but it is easy to lack overall planning and accumulate risks. In particular, the more fundamental a project in the DeFi ecosystem is, the more likely it is to introduce the risk of a single point of failure, even though it has a more "moat" status.

Many researchers organize the DeFi ecology according to different business types, but it is more necessary to draw an overall " risk map" for the DeFi ecology. In the future, DeFi projects should also conduct financial risk audits in addition to smart contract audits before going live.