This series is an in-depth discussion of DeFi issues. The previous article introduced the 7 major challenges that DeFi is currently facing. This article deeply discusses the liquidity problem, one of the biggest problems in DeFi, and introduces major projects in the DeFi field and their solutions to liquidity problems The method deeply analyzes the similarities and differences, advantages and disadvantages of its mechanism, which can enable readers to deeply understand the current status of the DeFi field and deepen the understanding of the industry.

This is a series of articles. The first part deeply analyzed the seven major challenges of DeFi. The title is "DeFi""Danger" and "Machine".

The content of this article will give an overview of blockchain-based asset transactions from the perspective of centralization/decentralization, focusing on the different methods used in terms of liquidity implemented by 0x, Injective Protocol, Kyber Network, Bancor, and Uniswap.

The most widely used application of blockchain technology is the creation of digital currencies, which requires the development of financial markets to support their transactions. However, these financial markets in their current state prevent normal use by the general public. The infrastructure that underpins blockchain asset transactions is also vulnerable to counterparty risk, censorship, lack of transparency, and manipulation because it remains very centralized.

The recent Binance theft incident is the presentation of these shortcomings;secondary title

1. Blockchain-based asset transactions are still centralized

On centralized cryptocurrency exchanges, custody of funds and/or order books is maintained by a third party. According to ConsenSys, as of late January 2018, an estimated 99 percent of cryptocurrency transactions were conducted through centralized exchanges, which are widely distributed.

secondary title

2. Centralized Exchange vs. Decentralized Exchange

While the advantages of centralized exchanges include efficiency, price discovery, liquidity, and market depth, there are also important disadvantages. Centralized exchanges are more vulnerable, less transparent (instances of manipulation), and prevent certain market participants from using them.

Security Holes: Hosting

The biggest vulnerability of centralized exchanges is that they keep custody of users’ assets. About 73% of cryptocurrency exchanges are custodial for user assets, meaning the exchange manages customers’ cryptocurrency wallets and keys. This creates a very attractive single point of failure for hackers when keys are held by exchanges.

This is a huge headache for a blockchain network; it is designed to eliminate such single points of failure. Notable examples of this include Mt. Gox and more recently the QuadrigaCX exchange. Last year, Gnosis stated, “A series of hacks that resulted in a loss of $1.5 billion on centralized exchanges hit users hard and greatly affected user confidence in cryptocurrencies.

"The main value proposition of most DeFi projects is that they don't require users to give up custody of their assets.

Lack of Transparency: Manipulation

Many centralized exchanges operate off-chain, which means that most transactions are not actually recorded on the blockchain. Instead, the exchange acts like an escrow fund for customers, which makes transparency less clear full. This can lead to money laundering transactions, buying and selling to artificially create transactions.

After analyzing 81 exchanges, Bitwise found such behavior on 71 and estimated that about 95% of all bitcoin transactions were faked by exchanges.Motives for faked trades vary, but are usually to indicate higher liquidity, as exchanges are evaluated based on these metrics.

The Bitfinex/Tether investigation highlights opportunities for overt manipulation on centralized exchanges. Data analysis from Mt. Gox shows that market manipulation is also prevalent on the exchange. Centralized exchanges also exert centralized power influence when deciding which tokens to list or which chain to support after a hard fork.

limited use

secondary title

3. Alternative: Decentralized Exchange (DEX)

To counter these issues, there has been a trend to create an exchange that is as decentralized and permissionless as cryptocurrencies themselves.More specifically, a decentralized exchange is a platform that does not require centralized custody of funds, or trust in a third party to facilitate transactions or settlements.

According to Delphi Digital, as of June 2019, the largest DEX (by trading volume) is IDEX, followed by Eth2Dai/Oasis DEX, Uniswap, and Bancor.

In fact, most "DEXs" are non-custodial exchanges. On a non-custodial exchange, users remain in control of their keys and thus retain ownership of their crypto assets. However, non-custodial exchanges still store off-chain trading orders on centralized servers, which requires a higher degree of trust than using centralized exchanges.

Typically, speed and cost are limiting factors for on-chain transaction execution, but running off-chain brings more centralization.

Somewhat ironically, regulation may push these exchanges to innovate faster to achieve greater decentralization. Following the EtherDelta ruling, in which the SEC accused founder Zach Coburn of violating federal securities laws and fined him $388,000, DEXs opted to pre-emptively keep their strategies out of third-party participation as much as possible.

However, the more decentralized exchanges become, the more complicated it is to create liquidity.Liquidity is not just a "phantom" metric, without which markets cannot function properly and assets will trade at a discount.Compared with centralized exchanges, decentralized exchanges have fewer users and a more decentralized trading ecosystem, which disperses liquidity and hinders arbitrage between exchanges.

To get an idea of ​​how fragmented both centralized and decentralized cryptocurrency exchanges are, CoinMarketCap lists 253 exchanges. Based on adjusted daily trading volume, the top exchanges hold only about 3.5% of the market (as of June 28, 2009).

Generating liquidity in a decentralized market is a complex problem, and often in the process of solving one problem another problem arises.

secondary title

4. Networked liquidity vs advance trading

Launched in August 2017, 0x is not a DEX per se, but a protocol that allows “Relayers” to create DEXs on a public smart contract system. To some extent, relayers are to 0x what DApps are to Ethereum. Importantly, relayers do not escrow users’ funds or process transactions, they simply broadcast orders for peer-to-peer transactions.

When creating a DEX on the 0x protocol, the relay party can make different design choices (order strategy, fee design, etc.), and the protocol creates a basic pool of networked liquidity that can be used by all relay parties, and Outline the design structure of the order. It utilizes an off-chain order book combined with on-chain settlement, similar to other non-custodial exchanges. As of December 2018, there were 16 relayers on the protocol and 19 projects using the protocol, including District0x, MakerDAO, and Dharma.

In the 0x model, relayers can choose to adopt either of two general order strategies: open order or matching order. The open order book approach facilitates networked liquidity across the 0x protocol, but is susceptible to front-running and trade conflicts.

In an open order, the "Maker" publishes the order to the network without specifying a specific "Taker". Relay parties aggregate these orders, display them on their order sheets, and allow participants to match the order. This will result in a transaction conflict where two "order takers" simultaneously eat the same order.

Conflicts can also arise if the taker processes the order while the maker cancels it. The system also has front-running, which occurs when one party is able to view the order flow before the order is processed, which is what happens in an open order book system. Front-runners profit by trading intermittently with higher transaction fees, thereby incentivizing miners to put their transactions into blocks before they are able to see the transactions in the order book.

The second approach is matched orders, which prevent front-running and trade conflicts, but do so at the expense of networked liquidity. With order matching, the "maker" specifies a specific "taker" address. This could be a relay party who then fills orders in bulk with overlapping prices, or a pre-designated "taker" address.

In the first case, users must trust that the relayer will honestly fill the order according to their specifications and not front-run the transaction. Reputational risk is the only incentive for relayers to act honestly if there are no draconian measures explicitly penalizing untrustworthy relayers when matching orders. In the second case, the taker is pre-designated, so the networked liquidity is not used.

Front-running is a problem on almost every network covered in this article, but it is an area that is constantly being researched and innovated. Research conducted includes splitting liquidity aggregation into two contracts (trade execution contract and trade execution coordinator), commit-reveal scheme, and using zero-knowledge proofs to aggregate orders.

0x is working with StarkDEX (the Starkware project) on the latter approach. StarkDEX aims to use zero-knowledge proofs to aggregate thousands of orders into one proof, which is then released to the chain, making front-of-the-box transactions more difficult and greatly improving scalability.

The business model of networked liquidity

In theory, the 0x system should lead to competition among relayers to create a better user experience, while the 0x protocol provides better liquidity.This model of creating other projects capable of building user-facing applications has become the de-facto adoption model for DeFi, with projects like Dharma, Set, and dYdX among others.

Although the benefits of networked mobility are logically obvious, they have not been proven effective in practice.

This concept requires exchanges to share liquidity, which requires a mind-shifting thinking from the conventional wisdom that the most liquid exchange is the “best exchange”. If the 2-3 relayers with the best user experience dominate, they may not have much incentive to share liquidity. If the majority of trades are executed on these advantaged platforms, then this will not gain much value from "network liquidity" at all.

However, the DEX and DeFi industries are still in their early stages of development. As the DEX industry matures, it should become differentiated. In doing so, it is more likely that different DEXs will emerge to facilitate the trading of specialized products.

secondary title

5. Converged liquidity vs price discovery

Converged Liquidity vs. Price Discovery: Kyber, Bancor and Uniswap.

Although networked liquidity brings together the liquidity of various DEXs built on the same protocol, other networks solve the liquidity problem from different angles.Converged liquidity pools liquidity brought in by market makers into a common “pool” that can be driven by reserve managers or incentivized at the protocol level.These methods are highlighted in detail below.

The Kyber network was launched in 2017. Its interface adopts a decentralized design and connects to different exchanges (including centralized ones), which enables on-chain transactions without implementing orders. Kyber pools the liquidity of different ERC-20 tokens managed by "reserve managers".Kyber can be thought of as a kind of reserve bank rather than an exchange, it reserves different cryptocurrencies, enabling instant transactions between ERC-20 tokens.

Other convergent liquidity networks solve problems at the protocol level (via smart contracts). These networks also do away with the concept of order books, but instead of relying on reserve managers, they rely on an automated market maker (AMM) model. AMMs pool liquidity and use deterministic algorithms to make markets by quoting traders based on predefined formulas, setting clear relationships between tokens. Price drops for large trades are an issue in AMM networks, as is front-running trades.

The amount to buy and sell on a network using an AMM is determined by a formula that uses an algorithm to determine the price, separating liquidity from volume, but also price from market forces.

This is not necessarily a problem; however, it is a matter of trade-offs. When using an AMM, users sacrifice efficient price discovery in exchange for guaranteed liquidity.

Even if the price is unfavorable, being able to trade may be preferable to not being able to trade at all in some cases.

Proponents of AMMs argue that AMMs eliminate the profits that market makers extract from the market and prevent market participants from behaving irrationally. In practice, this does not make the market more efficient. This is because these markets rely on arbitrageurs to enter and maintain a rough balance between open and automated markets. Consequently, these markets always lag behind open markets, but still reflect the irrationality of the participants.

This change became more pronounced with the launch of the Bancor network in 2016. The Bancor protocol sets prices according to the Bancor Formula. This formula sets the price at a pegged ratio of "Smart Tokens" to BNT (Bancor Network Token), which links all tokens together through "transitive connectivity". If this sounds like centralization, that's because it is.

Due to the design of its native token, Bancor is relatively expensive. Additionally, Bancor Formula relegates price discovery to a series of linear, incremental bids and offers, making it inefficient by design.

Launched in late 2018, Uniswap took a very different approach. Uniswap consists of a series of smart contracts deployed on Ethereum to facilitate the trading of ERC-20 tokens through an AMM version called "Constant Product Market Maker".

In simple terms, Uniswap creates a smart contract per token that creates a constant ratio between the ETH liquidity pool and the "connected" token liquidity pool. Unlike Bancor, this formula allows for fluctuations in the transaction ratio of ETH to ERC-20 tokens and incentivizes market makers in a way that facilitates transactions.

Specifically, 0.3% of the entire trading volume is distributed proportionally to all liquidity providers. These fees are returned to the liquidity pool until the market maker chooses to cash out, which increases how profitable the next trade is. The entire process takes place on-chain, using Ethereum as a common medium of exchange, which eliminates the need for native tokens and greatly reduces gas costs relative to Bancor.

Compound’s Robert Leshner described Uniswap’s advantages over Bancor this way,“Uniswap does away with native tokens and simplifies the algorithm. It only requires a fraction of gas to trade and incentivizes the community to continuously increase liquidity.”

There are still challenges in the Uniswap model. Still, it is pushing the industry in the right direction: decentralized usage, on-chain transparency, cleaner UI/UX, and clearer incentive design.

The business model of converging liquidity

Although Bancor and Uniswap compete directly, the relationship between Kyber and Uniswap has so far been mutually beneficial. While Uniswap provides ample liquidity for certain tokens, it has relatively few user-facing integrations.

Kyber, on the other hand, has struggled to generate sufficient liquidity, but facilitates the integration of a large number of DApps, wallets, and providers. Recognizing the complementary strengths of each platform, Kyber added Uniswap to its liquidity reserve in February 2019, which accounted for 34% of Uniswap’s total trading volume from February to May.

secondary title

6. Dream and reality

Most blockchain-based asset transactions are still centralized. Even within the DeFi industry, a centralized approach to fiat onboarding is still the first step into most markets (although there is innovation now), and most non-custodial exchanges are still centralized in many ways. While regulation may push the design toward more practical decentralization, given the complexity, this may not happen overnight.

At the same time, we may see hybrid designs catch on. At some point in the future, there may be an inflection point where traditional traders need to access assets often through Ethereum-based contracts, leading to a combination of two worlds. Until then, the DeFi industry will have to continue to work hard to create real liquidity, prevent front-running and other manipulations, and promote effective cross-chain transactions and cross-market arbitrage so that users don’t have to make too many trade-offs.