Cobo Private Meeting: The FTX Incident Marks the Dawn of the Centralized Exchange Era

Author: Cobo

The crypto industry seems to be facing a twilight moment this year: from the fall of Luna to the collapse of 3AC, and then to the downfall of the FTX empire, a series of negative events has cast a shadow over the development of the entire industry.

Will you still believe in Crypto?

In this tumultuous autumn, blindly recharging faith is not a wise choice; it is more important to learn from everything that has happened and make reasonable judgments about the future of the industry.

Recently, Cobo held a private meeting where Cobo CEO & Co-Founder Shen Yu and other senior practitioners in the field expressed their analyses and views on events like FTX. The themes covered included interpretations of multiple black swan events, changes in the decision-making processes of centralized institutions, and the future development of the market, all of which are worth our reference and reflection.

Deep Tide has edited and organized Shen Yu's insightful points from the private meeting, filled with valuable content to share with readers.

Three Black Swans Stirring the Twilight of Exchanges

In 2022, the crypto circle faced a significant turning point.

The three major black swan events of Luna, 3AC, and FTX had destructive power and influence far beyond previous years. Tracing back to the source, we find that the crisis had quietly laid its groundwork much earlier: the FTX incident can be traced back to the collapse of Luna, and recent internal documents have confirmed that many of FTX's losses originated from earlier times.

Observing the source, those familiar with DeFi principles can understand that this year's rapid collapse of Luna was a typical Ponzi scheme: an abnormal market event, a rapid bank run, and in an instant, the hundred billion dollar market cap of Luna vanished. In this event, many centralized institutions were inadequately prepared for market expectations, leading them to have significant risk exposures, such as 3AC quickly transforming from a risk-neutral hedge fund into a one-sided gambler.

In the subsequent timeline, crises unfolded one after another.

In June, many institutions held asymmetric positions, using high leverage to go long on Bitcoin and Ethereum, blindly believing that certain fixed points would not be breached, leading to mutual lending among institutions, which resulted in the 3AC incident; by September, with the completion of Ethereum's Merge, the market began to show signs of recovery, but an unexpected event triggered the collapse of FTX.

Regarding the FTX incident, from CZ's perspective, it might be seen as normal business competition, a targeted attack on a competitor's financing. However, the situation unexpectedly led to market panic, ultimately revealing Sam's massive financial hole, which then triggered a rapid panic sell-off, resulting in the swift downfall of the FTX business empire.

In this year's three major black swan events, several significant yet often overlooked points are worth our deep reflection:

First, institutions can also go bankrupt.

Especially large institutions in the Western world, after a massive influx of institutional users in 2017, the entire industry became highly correlated with the U.S. stock market.

For retail/individual investors, these institutions seem mysterious, bringing in substantial funds and capabilities. However, from this year's black swan events, it appears that many North American institutions had some misconceptions regarding risk management or their understanding of the entire Crypto world, leading to numerous incidents and a chain reaction among many institutions.

Therefore, our conclusion is that institutions can go bankrupt and restructure; the uncollateralized credit among institutions is highly transmissible.

Second, quantitative and market-making teams can also incur significant losses in extreme events.

During rapid surges and drops in the market, especially during downturns, the market becomes filled with distrust towards institutions. This leads to a massive outflow of funds and severe liquidity shortages, causing many market-making teams to passively convert their high liquidity assets into low liquidity assets, facing lock-up and inability to withdraw.

In multiple rounds of black swan events, several such market-making/quantitative teams were affected.

Third, asset management teams will also face impacts.

Asset management teams need to find low-risk or even risk-free interest rates for a large amount of assets in the market to provide returns to investors. The ways to achieve alpha returns essentially boil down to two: lending and token issuance.

The former generates returns by providing liquidity to the market, while the latter issues tokens to the market through consensus mechanisms (like PoW, etc.), ICOs, and DeFi early mining. During the operation of asset management teams, a large amount of lending assets and related derivatives accumulate. Once a black swan event occurs, such as an institutional collapse, the lending assets will trigger a chain reaction, facing significant shocks in extreme market conditions.

This inevitably brings to mind traditional financial markets.

The crypto market has developed rapidly, seemingly completing the process of traditional financial markets over more than 200 years in just over the past decade, with both excellent stories and recurring issues in traditional financial history. For instance, the behavior of banks hollowing out commercial loans also occurred in the FTX incident. And everything seems to point to the operational issues of centralized institutions.

At the same time, the FTX incident also basically marks the arrival of the twilight of centralized exchanges. Globally, there is extreme panic regarding the opacity of Crypto, especially CEX, and the potential chain reactions it may cause. Data also confirms this judgment, as there has been a significant amount of asset transfer activity on-chain in the past month.

Before the twilight, private keys lost the battle against human nature:

In the Crypto world, the ownership of underlying assets is guaranteed by private keys, but in the past decade of development, centralized exchanges have not had a reasonable third-party custodian to help users and exchanges manage assets, thus combating the human issues of exchange managers, allowing exchanges to always have the opportunity to touch users' assets.

In the FTX incident, the influence of human nature seems to have been traceable.

Sam has always been a person who cannot sit idle. Various forms of overtime and late-night work prevent him from allowing himself and money to remain idle. During DeFi Summer, it was also observed that Sam frequently transferred large amounts of assets from the exchange's hot wallet to various DeFi protocols for early mining.

When human nature craves more opportunities, it also struggles to resist more temptations.

A large amount of user assets placed in the exchange's hot wallet, using assets to obtain no (or low) risk returns seems logical; from staking to mining DeFi, and then to investing in early-stage primary market projects, as returns grow larger, misappropriation may escalate.

The black swans have turned the industry upside down, leaving us with an obvious lesson: for regulators and large institutions, they should learn from traditional finance, find appropriate ways to ensure that CEX is no longer a single entity simultaneously bearing the roles of exchange, broker, and third-party custodian; at the same time, technical means are needed to make third-party custody and trading behaviors independent, ensuring that interests are unrelated. If necessary, regulatory intervention may even be required.

Outside of CEX, other centralized institutions facing industry upheaval may also need to make changes.

Centralized Institutions: From "Too Big to Fail" to Rebuilding with Purpose

The black swans have stirred not only CEX but also industry-related centralized institutions. A significant reason they were affected by the crisis is that they overlooked the risks posed by counterparties (especially CEX). The impression of FTX was that it was "too big to fail." This was also the second time Shen Yu heard this concept: in some group chats in early November, most people voted that FTX was "too big to fail."

The first time was when Su Zhu personally told Shen Yu: "Luna was too big to fail; if it fell, someone would come to save it."

In May, Luna collapsed.

In November, it was FTX's turn.

In the traditional financial world, there is a last resort lender. When large financial institutions experience severe events, there are often third-party organizations, even government-backed organizations, that conduct bankruptcy restructuring to mitigate the impact of risks. Unfortunately, there is none in the Crypto world. The entire Crypto world is transparent at its core, and everyone analyzes on-chain data through various technical means, leading to rapid collapses. A small clue can trigger chaos.

This phenomenon is a double-edged sword, with both advantages and disadvantages.

The advantage is that it accelerates the bursting of bad bubbles, allowing things that should not happen to disappear quickly; the disadvantage is that it leaves little opportunity window for less sensitive investors.

In this market development process, Shen Yu maintains the previous judgment: the FTX incident basically marks the arrival of the twilight of centralized exchanges, in the future, they will gradually degrade to become a bridge connecting the fiat world and the Crypto world, and solve issues like KYC and deposits through traditional means.

Compared to traditional methods, Shen Yu is more optimistic about on-chain operations that are more open and transparent. As early as 2012, there were discussions about on-chain finance in the community, but at that time, it was constrained by limited technology and performance, lacking suitable carrying means. With the development of blockchain performance and underlying private key management technology, decentralized finance on-chain, including decentralized derivatives exchanges, will gradually rise.

The game has entered the second half, and centralized institutions also need to rebuild in the aftershocks of the crisis. The cornerstone of rebuilding remains the ownership of assets.

Therefore, in terms of means, using the currently popular MPC-based wallet technology solution to interact with exchanges is a good choice. Institutions should retain ownership of their assets and then ensure the safe transfer and trading of assets through third-party cooperation and exchange agreements, allowing transactions to occur within a very short time window to minimize counterparty risks and chain reactions triggered by third parties.

Decentralized Finance: Finding Opportunities in Crisis

As CEX and centralized institutions suffer, will the situation for DeFi be any better?

With a massive outflow of funds from the entire crypto world and the macro environment facing interest rate hikes, DeFi is facing significant shocks: looking at overall yields, DeFi currently lags behind U.S. Treasury bonds. Additionally, investing in DeFi also requires attention to the security risks of smart contracts. Considering both risks and returns, the current situation of DeFi appears rather pessimistic in the eyes of mature investors.

In this relatively pessimistic environment, the market is still brewing innovation.

For example, decentralized exchanges focusing on financial derivatives are gradually emerging, and innovations around fixed-income strategies are rapidly iterating. As the performance issues of public chains are gradually resolved, Shen Yu is also optimistic that the entire DeFi interaction methods and possible forms will undergo new iterations.

However, this update and iteration will not happen overnight; the current market is still in a relatively delicate phase: due to black swan events, crypto market makers have suffered losses, leading to severe liquidity shortages in the entire market, which also means that extreme cases of market manipulation occur from time to time.

Assets that were previously liquid are now very easily manipulated; once price manipulation occurs, due to the existence of numerous combinations among DeFi protocols, many entities may inexplicably be affected by price fluctuations of third-party tokens, leading to liabilities.

In such a market environment, corresponding investment operations may become conservative.

The Shen Yu team currently prefers to seek more stable investment methods, obtaining new asset increments through staking. At the same time, they have developed a system called Argus to monitor various on-chain anomalies in real-time, improving overall operational efficiency through (semi) automation. As industry OGs gradually adopt a cautiously optimistic attitude towards DeFi, we are also curious about when the entire market will see a turning point.

Anticipating Market Reversal: Both Internal and External Factors Are Essential

No one enjoys a crisis indefinitely. On the contrary, we all look forward to a turning point. But to predict when the wind will change, we must first understand where the wind is coming from.

Shen Yu believes that the previous round of market fluctuations likely stemmed from the entry of traditional investors in 2017. Due to the relatively large asset volumes they brought, combined with a relatively loose macro environment, a heated market was created. Currently, perhaps the bear market will see a reversal only when interest rates are lowered to a certain extent, allowing hot money to flow back into Crypto.

Additionally, in previous rough estimates, Shen Yu believes that the total daily costs incurred by the entire crypto industry, including mining machines and practitioners, is approximately tens of millions to one hundred million U.S. dollars; however, the current on-chain capital flow situation shows that the daily capital inflow is far less than the estimated expenditure, indicating that the entire market is still in a stage of stock game.

With liquidity tightening and stock game dynamics, the poor external environment can be seen as an external factor preventing the market from reversing. The internal factor that can drive the crypto industry upward comes from the growth points brought about by the explosion of killer applications.

Since many narratives have gradually quieted down after the last bull market, the industry has yet to clearly identify new growth points. As layer-two networks like ZK gradually launch, we vaguely sense changes brought by new technologies, with public chain performance advancing further, but we have not yet seen clear killer applications; reflecting on the user level, we still do not know what application forms will lead to the influx of assets from large-scale ordinary users into the crypto world. Therefore, the end of the bear market has two prerequisites: one is the lifting of external macro interest rate hikes, and the other is finding the next new growth point from the explosion of killer applications.

However, it is important to note that the reversal of market trends must also align with the inherent cycles of the crypto industry. Considering the Ethereum Merge event in September this year and the upcoming Bitcoin halving in 2024, the former has already occurred, while the latter is not far off from an industry perspective. In this cycle, there is actually not much time left for breakthroughs in applications and narrative explosions within the industry.

If the external macro environment and internal innovation pace do not keep up, then the existing understanding of a four-year cycle within the circle may be broken, and whether the bear market will become longer across cycles remains to be observed and learned. When both internal and external factors that can prompt market reversal are essential, we should gradually accumulate patience and adjust our investment strategies and expectations accordingly to face more uncertainties.

Things have never been smooth sailing; may every participant in the crypto industry be a solid Builder rather than a bystander missing opportunities.

Appendix: Highlights from the AMA Q&A

Question 1: What are the main innovation directions for the future of the crypto market?

The major tracks for the future of the entire crypto market are two, or in other words, the entire industry has basically developed around two things over the past decade:

The first thing is performance, specifically the issue of TPS. From the expansion in 2017 to now, the major direction still requires multi-layer networks to solve it. Currently, the zk possibilities within layer-two networks are the most promising, but the timeline for final implementation and usage may still be quite long, possibly at least two years.

The second core direction is the balance between the security of underlying private keys in the entire Crypto world and their application, which is also a long-standing issue that has always hindered the emergence of a large number of new users in the industry. With the influx of traditional capital and a large number of new users over the past five years, whether in GameFi or applications like StepN, MPC-based keyless wallets may provide a more balanced user experience and threshold. So, in the long run, these two major directions are the core issues that the entire industry must solve.

Question 2: How do you view the current market situation, and how will it develop in the future?

The entire industry is currently in a state of stock game, as the outflow of assets is still quite severe, so it is undoubtedly a bear market, and the prices have dropped significantly. Comparing previous rounds of declines, asset retracements have been around 80%, so it is still very difficult to determine when the bottom will be reached. Even if it is not the absolute bottom, it is still within the bottom range; how long this period will last, what state it will reach, and when the next turning point will occur all require continuous observation rather than impulsive decisions.

Therefore, I believe there are two potential major turning points. The first is the end of the entire interest rate hike cycle.

From a historical perspective, the shortest interest rate hike cycle over the past few decades has lasted more than a year. If we start counting from March 2022, it will likely last until mid-2023. From an industry perspective, there definitely needs to be a new growth point and trigger to bring new capital inflows, and this needs to be continuously observed to determine the range. However, whether there will be further declines or more black swan events remains uncertain. But it is undeniable that the prices are already relatively low. From another indicator, the perspective of miners shows a typical signal of a bottoming cycle, as the mining costs are no longer able to cover marginal costs, and they can hardly afford electricity bills.

However, I want to remind everyone that the miners in this market have experienced the mining ban of 2021, so 80% of miners are currently in developed countries, and their mining agreements often do not allow for flexible shutdowns. Therefore, many miners may not choose to shut down their operations even under these circumstances, so the hash rate may only see minor declines; last week's data showed a slight decrease in mining difficulty.

Additionally, the unique situation of North American mining farms is that many miners made mining machine loans and unsecured loans with many lenders during the previous bull market cycle. Currently, while we do not see a large-scale sell-off of mining machines, we have observed two interesting phenomena: first, a large number of mining farms are undergoing bankruptcy restructuring, which is a characteristic action in the West. Second, the second-hand market price of mining machines has likely dropped from several tens of dollars per terahash to around ten dollars per terahash, already below the production cost price.

So from the perspective of miners and production supply, they are already in a surrender phase, so I definitely believe this is the bottom range. However, how long the bear market lasts may still require observation for a turning point. For individual investors, this is a very good opportunity because we can see that these institutional investors are already underwater, and many have even gone bankrupt.

Often during such times, when the market is in turmoil, if you still have cash reserves that do not significantly impact your life, you can gradually build positions over time. In the long run, this should be okay; this cycle may be shorter, around one to two years, or longer, possibly three to five years.

Question 3: Signature Bank is selling $10 billion in crypto deposits. Is this a landmark event that will cause a tightening of stablecoins in the crypto market? How do you view the impact of Signature Bank and Silvergate from FTX?

The first question is that the outflow of stablecoins will definitely be a landmark event, so everyone can check the issuance volume of circulating stablecoins in the industry weekly. A large outflow will indeed put significant liquidity pressure and price shocks on the entire industry.

As for the second question, I personally feel that these two banks have limited connections with FTX based on the current public channels and data, and their degree of impact is relatively limited. After all, they are more traditional, and although they have some lending business, the data appears to be relatively good.

For these banks, as cryptocurrency investors, we normally would not use them as private banks. They are merely used for account opening and as fiat entry and exit points, so there is no need to hold large amounts of assets there.

Question 4: If CEX wants to continue to grow in the future, what position will they hold in the industry, and how will their development look?

This question can be viewed from two angles.

First, due to the collapse of FTX, institutional investors have faced severe repercussions, and future regulations will certainly tighten, especially for centralized exchanges. Therefore, the historical model of centralized exchanges combining multiple functions may be dismantled in the future, which is an inevitable process in the development of traditional financial markets. From this perspective, I believe the current unified state of centralized exchanges may change.

The second perspective is that with the development of other decentralized exchanges and derivatives, competition may become greater, and users, influenced by these events, will be educated and have clearer awareness. In this situation, centralized exchanges will play a crucial core role, developing a set of solutions that match traditional finance, allowing these investors to learn, understand, and recognize the underlying assets that make up the entire Crypto world. Another function is that they will serve as the core channel for deposits and withdrawals, facilitating smoother interactions between the entire Crypto world and fiat currencies. Therefore, I believe these should become two very important roles for CEX in the future.

Their current core trading functions, especially in derivatives trading, will coexist with decentralized exchanges and may even gradually be replaced.

Question 5: Is there a possibility that Grayscale will collapse?

Today's bankruptcy restructuring or current difficulties may not trigger a chain collapse for Grayscale, as these two companies should be relatively well-isolated, and based on known SEC public data and Coinbase's disclosure of Grayscale's custody data, the assets managed by Grayscale should be safe unless significant unknown factors arise.

Additionally, from a business perspective, Grayscale is also a very good cash cow. If Genesis's parent company, DCG, faces significant liabilities, they may consider divesting Grayscale; meanwhile, companies and consortiums with substantial cash reserves may recognize Grayscale and begin acquisitions.

So based on known data, I believe the likelihood of Grayscale collapsing is low. However, its secondary market negative premium may persist for a while due to liquidity issues and market fears, so if some colleagues are considering arbitrage in the secondary market, I think it is possible but should be approached with caution, as this cycle may last longer.

Furthermore, regarding predictions of chain reactions, it is hard to say who else might collapse next. From the current known information, there is not much visibility, as centralized entities are relatively complex and operate as a black box, requiring time and effort to analyze their address relationships and asset transfer situations to uncover the underlying logic.

Currently, the entire market may still have some undisclosed events fermenting and expanding, but the more malignant and larger-scale issues have likely passed. There may still be some marginal or smaller-scale issues, like the example of 3AC, which was a ripple effect following the Luna incident.

In fact, if there are small exchanges that collapse in the next month or two, it will also be based on the fermentation of the FTX incident, and within about one to two months, some long-tail associated companies will gradually be exposed.

Question 6: What is the Cobo MPC solution like? How does it match market needs?

Cobo has always focused on the management of underlying private keys and the handling of on-chain risks. Since 2018, we have been dedicated to this area for three to five years, during which we have been contemplating the ultimate form of private key management and possible on-chain interactions, as well as internal risk management among organizations. After more than two years of thinking, our strategy has gradually become clearer, which we internally refer to as the Cobo three-step strategy:

The first step is to provide a completely centralized solution called Cobo WaaS (Wallet as a Service), which packages the management of underlying private keys interacting with the blockchain, chain access, monitoring of abnormal chain states, such as hard fork monitoring, into a standard service. This allows institutional users to access dozens of public chains through a single API, supporting many tokens conveniently for secondary development of their own businesses. This was our initial pure centralized solution.

We have also developed a Loop clearing network on this basis, addressing the core issue of preventing unnecessary construction in the wallet direction. At the same time, we aim to provide a standardized interface that allows everyone to easily access various underlying blockchains to utilize the functions of receiving, storing, and transferring assets on the blockchain.

In the second phase, as blockchain technology develops and we enter DeFi summer with more on-chain interaction scenarios, in addition to the initial functions of receiving, storing, and transferring on-chain, we have developed an early version of Cobo Argus driven by internal needs, which essentially enables a team to collaboratively manage on-chain assets with multiple people and permissions. In the Crypto world, I have used a multi-signature solution for quite some time, which essentially requires equal rights among multiple administrators, with every operation needing to be reviewed by multiple people, greatly reducing team collaboration efficiency.

So in the on-chain version, we use smart contracts to grant different permissions to different individuals. For example, some can manage funds for receiving and transferring; others may only hedge specific tokens, conducting trades to hedge certain risk exposures, possibly with speed limits and flow controls. Other operators may claim certain profits…

We have defined different roles for various scenarios, allocating different permissions, and conducted a series of encapsulations for the management of various DeFi interaction assets on-chain, allowing teams to match various internal needs while managing risks across protocols, supplemented by monitoring data dashboards.

In the third phase, once we can solve blockchain performance issues, we will form a multi-chain, multi-layer state. At this stage, we aim to leverage Cobo's current technological accumulation in MPC, and after iterating for about half a year, we have developed a Cobo Chain—a decentralized custody application chain. We drive these MPC nodes through blockchain consensus, allowing everyone's private keys to be distributed globally from day one, managing workflows and various risks through programmable internal workflows.

The above may be a bit further out, but currently, due to the outbreak of the FTX incident, institutions have a heightened demand for asset security and private key management. We have quickly packaged the MPC underlying technology accumulated on the application chain into Cobo's MPC WaaS version and are currently communicating with institutional investors to match various internal refined customization needs, standardized MPC needs, super loop clearing networks, etc. This area is currently being iterated and refined, and the ultimate state may be a decentralized custody chain that can match the needs of different blockchains from day one.

At the same time, we can also embed some standardized financial scenarios and models into the application chain, such as integrating some models that have been very mature in the traditional world into the entire chain's native layer. Currently, this is still in the early stages, and we are brainstorming and prototyping internally, and we welcome everyone to engage in discussions.

Question 7: Will there be such huge black swans appearing again in the future?

Currently, I do not see particularly large black swans, but aftershocks will definitely continue, as this year's three black swan events involved asset scales in the billions of dollars. Given the density and interconnections of their outbreaks, they have exceeded expectations. In the entire crypto world, there are not many well-known institutions that are still surviving, with only a few remaining. Based on known information, there are no significant risk exposures or issues. However, under tightening regulations, there may also be new rounds of shocks, which will need to be observed.

Question 8: Is now a relatively good entry point, and why?

First of all, we are currently at the bottom range of the bear market, but how long the bear market lasts is unknown. The first point is to maintain good cash flow; you need to sort out your cash flow assets and future situations, possibly making relatively pessimistic expectations.

In a stable cash flow situation, when survival pressure is not particularly challenging, I think it is possible to gradually build positions in some major asset classes, such as Bitcoin and Ethereum, after today or after March next year, especially for new protocols and projects that remain active and grow during the bear market. The current valuations are relatively much lower than during the bull market, so gradually building positions is advisable.

Additionally, observe whether there are new growth points and the emergence of new killer applications in the industry. If there are potential opportunities, it would be wise to invest a portion of assets at a reasonable price and low cost, which could yield better returns in the next round.

Question 9: How should institutions prepare to enter the DeFi track?

I believe DeFi has four major directions: the first direction is stablecoins, and there will soon be some interesting stablecoin developments, such as those from AAVE and Curve.

The second is lending. After the market stabilizes, lending will follow. Currently, after experiencing extreme conditions, on-chain lending has performed relatively well but has also exposed some issues, so there will likely be iterative thoughts to address some abnormal asset isolation problems. We are also observing the next developmental phase of lending.

The third is DEX, which has matured after a long development and was a strong driving force during the last DeFi Summer. Now DEX is relatively mature, and with the launch of ZK, the order book model may gradually improve.

The fourth is derivatives and risk management, which is relatively early, but after the FTX collapse, the volume of on-chain futures and options derivatives has started to rise significantly, so I believe this could be a new growth point in the near future.

For institutions, the first step is to establish awareness, and then to create an internal workflow tailored for DeFi. For example, Cobo Argus is a decentralized interaction tool that can meet institutional needs, which we have refined based on our internal requirements and can be said to be unique in the market.

For individual investors, if you want to obtain relatively stable returns in DeFi now, it is relatively difficult, so individuals should focus more on staking single tokens. These returns may be modest but carry lower risks, and individuals can experiment with small amounts of capital, observing new growth points and directions. There should be one or two promising growth points in the next round, so if you can understand these new developments during this bear market, it will greatly help in capturing growth and wealth in the next round.

Especially when we look back, many stories that occurred in the summer of 2018 with EOS were basically replicated in 2020/2021. Therefore, by revisiting the stories from the last bear market, we can anticipate some signs and reduce trial and error in the next round. During this phase, it is advisable to observe more, act less, learn more, and read more.