VC Perspective: Hyperliquid Event Unveils the Power Struggle Between CEX and DEX

Original Title: Exchange War Erupts: Hyperliquid vs. Binance & OKX - The Chopping Block

Original Source: Unchained

Original Compilation: Deep Tide TechFlow

Guests:

• Haseeb Qureshi, Managing Partner at Dragonfly
• Robert Leshner, CEO & Co-founder of Superstate
• Tarun Chitra, Managing Partner at Robot Ventures
• Tom Schmidt, General Partner at Dragonfly

Highlights of This Episode

The JELLYJELLY Crisis of Hyperliquid - How a prominent DeFi project lost market trust by using false oracle prices to rescue its treasury.

Escalation of Competition Among Trading Platforms - The listing of JELLYJELLY perpetual contracts on Binance and OKX is interpreted as a precise strike against Hyperliquid.

Are Decentralized Exchanges Really Decentralized? - The Hyperliquid incident reveals issues arising from the concentration of validator power behind the so-called "decentralization" performance.

DNA on the Blockchain: The Blockchain Version of 23andMe - Say Foundation proposes to protect genetic data through token thresholds; is this an innovation in privacy protection or a dystopian vision?

The Scam of Decentralized Science (DeSci) - Tarun criticizes the DeSci concept again and explains why the risks of putting genetic data on-chain are more severe than meme coins.

The Regulatory Battle Over Stablecoins - The "Stable Act" and "Genius Act" are competing in Washington; who will come out on top?

The Possibility of Stablecoins as Narrow Banks - The rise of cryptocurrencies may force the Federal Reserve to accept a financial concept they have resisted for 20 years.

The Future Gamble of HLP Deposits - The host bets real money on whether the deposit volume will rebound or continue to decline after Hyperliquid's collapse.

New Developments in Memecoins and Olympus - Are those once considered "robbers" quietly profiting from the shattered treasury?

Tarun's Failure Rankings - Why the failure of JELLYJELLY is worse than MobileCoin, but at least it aligns with Hyperliquid's brand positioning.

The Hyperliquid Incident Unfolds

Haseeb:

One of the biggest stories this week is the drama unfolding on Hyperliquid. For those unfamiliar, Hyperliquid is a new hot DEX, now the number one DEX by overall trading volume. They conducted a massive airdrop, which was well-received by crypto retail investors due to the scale of the airdrop and the fair launch method.

In recent days, Hyperliquid faced a large-scale attack surrounding a memecoin—JellyJelly, which is a low liquidity and peaked memecoin, but was listed as a contract trading option on Hyperliquid. One trader opened a short position worth $8 million on JellyJelly, which was equivalent to 50% of Jelly's circulating market cap at the time, a very large short position. Subsequently, this trader drove up the spot price of Jelly, leading to their forced liquidation.

So, why did the trader do this? Why force themselves to be liquidated?

On Hyperliquid, when a position cannot be liquidated in the normal way, HLP (Hyperliquid's crowd-sourced market makers) will take over the position and attempt to liquidate it in an orderly manner. The existence of HLP is crucial for Hyperliquid's liquidity, as it always provides liquidity to traders. However, due to the massive scale of this position, HLP was forced to short Jelly, but no one in the market was willing to take on this short position, ultimately leading to what is known as a "short squeeze."

This short squeeze was not merely a retail investor's prank. In fact, the two major trading platforms, OKX and Binance, were also indirectly involved. When the market realized that Hyperliquid or its HLP was massively shorting Jelly, OKX and Binance announced they would list JellyJelly's contract trading within 24 hours.

Almost everyone believed, "This is a war between trading platforms." CZ and the boss of OKX had their guns aimed at Hyperliquid, and this was their opportunity to strike.

The validators of Hyperliquid voted to delist JellyJelly and forcibly liquidated the position at a price below the market price by artificially adjusting the oracle price.

Haseeb:

Hyperliquid's decision sparked widespread controversy. They believed that rather than letting the platform or HLP holders bear the losses, it was better to artificially intervene and lock the price at a false low to protect the interests of HLP users. However, this move caused Hyperliquid's token price to plummet from about $21 to $15, with a nearly 25% drop in a single day.

This incident raised two core questions:

First, was Hyperliquid's response reasonable in this situation? Does their mechanism design have fundamental flaws?

Second, does this expose that Hyperliquid's level of decentralization is not as high as it claims?

These questions sparked intense discussions in the industry, with some centralized trading platforms (like Bitget) openly criticizing Hyperliquid's practices as unfair. The competition among decentralized trading platforms has also intensified, and the DeFi space seems to be experiencing a significant watershed moment.

So, what do you think about this HLP incident?

Tarun:

I think, this incident indeed exposed some flaws in the protocol design. Just like automated market makers (AMMs), the AMM mechanism does not allow for order rejection. For example, the initial Unicorn v2 and v3 had no flexibility in this regard; you cannot choose to accept or reject specific orders. This issue also exists in Hyperliquid's liquidity pools.

Hyperliquid's HLP mechanism differs from other platforms (like GMX's GOP and Jupiter's JLP). The operational logic of HLP is that users deposit Ether (ETH), and the platform allocates these ETH across multiple assets for market making. For instance, they might use 1% of ETH to market make JellyJelly, 90% for Ethereum, and the remaining portion for Bitcoin. These asset allocations are determined by off-chain algorithms, and users need to trust the Hyperliquid team's asset allocation capabilities.

Clearly, they made some mistakes in mechanism design, such as not setting position limits or caps on open contracts. If there were these limits, the issues could have been mitigated without the need for urgent intervention. Hyperliquid has indicated that they will fix these mechanisms, including adding limits on open contracts and concentration limits.

This is precisely why I mentioned that liquidity pools that do not differentiate order types have certain limitations. Under the current mechanism, HLP cannot selectively handle orders, meaning it cannot reject certain orders while only accepting specific types of orders. If HLP could differentiate between positions that were forcibly liquidated by third parties, the market could price these positions based on their actual value, and HLP would not have to bear unnecessary losses. However, the current design forces HLP to automatically trade with these positions, which operates similarly to how Unicorn pools function. Therefore, they lack sufficient restrictions in strategy design. These strategies are actually run off-chain by the Hyperliquid team, rather than being executed transparently on-chain.

I am not clear on what their code implementation looks like, as most of the code is not open source. While I can run a node, I only receive binary files and cannot view the source code. Additionally, many of the system's settings are not clear or transparent enough. This incident clearly indicates that they have significant flaws in strategy limitations. I believe this is an issue they acknowledge needs to be prioritized for repair. But from a market perspective, it also highlights the value of having more open strategies rather than being completely closed off as it is now. Because currently, the operational mechanism of HLP has almost no transparency to the outside world.

As an HLP depositor, you actually do not know whether HLP has clear risk limits, such as whether it will automatically bear the position risk of the entire liquidity pool. You also cannot determine whether they will intervene in the market by artificially adjusting oracle prices as they did in this incident. Although some related content is mentioned in the documentation, users cannot verify the true operation of these mechanisms since the code is not open source. Even without open-sourcing the code, there is a lack of other verifiable proofs to confirm its behavior.

I believe that the mechanism guarantees provided by Hyperliquid differ from the transparency users expect in other protocols. In other protocols, users can clearly understand the operational logic of the strategies, although this transparency may require some trade-offs in efficiency and flexibility. However, Hyperliquid's strategies are not public, which indeed improves the efficiency of capital utilization but also undermines user trust. This trade-off is not entirely wrong, but it is clear that they made less than ideal choices in certain decisions. However, these issues are understandable and can be fixed.

Controversy Over the Rescue Decision

Haseeb:

Is it reasonable to rescue HLP depositors? Clearly, HLP may face huge losses in this incident. Do you think this was a wrong decision?

Robert:

I think it was a mistake, to be frank. Trying to solve market issues after the platform's risk parameters have gone out of control is one thing, but liquidating positions through artificial intervention to profit HLP pools is inappropriate. Because in the contract market, one party's profit usually means another party's loss. In this case, the Hyperliquid team and validators seem to have mistakenly chosen who the winners and losers are.

HLP liquidity providers should bear the risk. If the liquidation is successful, they profit; if it fails, they accept the loss. However, this liquidation operation allowed HLP to profit, which also means other market participants suffered losses. I believe this violates the principle of market fairness. If they had to intervene in the price, they should at least not choose a price that benefits themselves. Even more confusing is that the price they chose was even lower than the market price at the start of the incident, clearly to make themselves the winners.

Tom:

I agree. This behavior also complicates the relationship between HLP as a product and the Hyperliquid platform as a whole. HLP is just one of the funding pools; users can also choose other funding pools and run different off-chain strategies. HLP is positioned as Hyperliquid's "official funding pool," but theoretically, anyone can create a funding pool. Therefore, most people would not assume that HLP would enjoy any special treatment. However, this incident made it seem like HLP received some preferential treatment.

Some people compare this incident to the "socialization of losses" or "automatic deleveraging" mechanisms of traditional contract trading platforms, but these two are not the same. In traditional mechanisms, when the market as a whole falls below the margin level, trading platforms freeze positions and spread losses across the insurance fund. In this incident, the loss status of HLP was merely compensated through artificial intervention, and Hyperliquid itself did not face default risk. This raises the question, why does HLP get to be the priority liquidity provider for the trading platform? If HLP fails to operate, why are they rescued?

Robert:

And the fact that it was done in a profitable way is just crazy.

Tarun:

Indeed. And ironically, HLP holders decided the price of this rescue through governance voting, indirectly bringing profit to themselves.

Robert:

Can you elaborate? How do HLP holders participate in the Hyperliquid validation process?

Tom:

HLP holders can delegate their voting rights to a validator, but some validators need to go through KYC, so the mechanism is a bit complex.

Tarun:

Validators control the oracle prices, so they decided on the price adjustments through governance voting. In other words, HLP holders actually participated in the voting indirectly through governance.

Haseeb:

Yes, this has received a lot of criticism. Because the Hyperliquid Foundation holds an absolute majority of voting power in HYPE tokens, in this case, HYPE holders quickly completed the voting through delegation. However, this entire process took only two minutes, from the start to the end of the voting, and the so-called voters had almost no real say.

Robert:

I'm still a bit confused about whether the listing of contracts on other trading platforms will affect Hyperliquid. The contract market and the spot market are relatively independent trading venues. Even if the demand for JellyJelly contracts on Binance increases, it may not directly change the prices of Hyperliquid or the spot market, as the spot price is controlled by the index, which also determines Hyperliquid's funding rates. So, what is the specific mechanism of this influence?

Haseeb:

First, if Binance wants to launch a spot market, they need to procure actual spot inventory, which takes longer to complete. But launching a contract market is faster than a spot market because the contract market does not require actual spot inventory. As long as there is sufficient demand, users can start trading contracts without immediately adjusting the spot price.

Robert:

Every long position corresponds to a short position, and every long has a short.

Haseeb:

Exactly. But structurally, starting a contract market is simpler. If you say, "Hey, I want to hit these guys quickly, and time is limited," then the fastest way is clearly through contracts, not spot.

The purpose of launching the contract market is to involve more people in the short squeeze. If Hyperliquid is experiencing a short squeeze, the opening of the contract market will exacerbate this trend.

Tarun:

This mainly depends on the dynamic changes in funding rates. In this incident, the funding rate skyrocketed by 300% in a short time, causing extreme market instability.

OKX and Binance Join In

Haseeb:

The funding rate soared by hundreds of percentage points. This was a very intense short squeeze, so the market expected this situation to calm down quickly. I suspect that Binance and OKX might delist the JellyJelly contract within a week or two, as it is clear that there is no real demand for this product.

Tarun:

No one will really need JellyJelly.

Haseeb:

However, I find it interesting that the mechanisms of this incident may be difficult to understand, especially if you are not familiar with the contract market, liquidity providers, and HLP operations. Let's simplify it first. The essence of the matter is that Hyperliquid got caught in a high-risk position, while Binance and OKX attempted to further weaken Hyperliquid's position through market operations. More specifically, their goal was to force HLP, rather than Hyperliquid itself, into insolvency.

This behavior is very aggressive, right? Some people compare this incident to CZ's actions against FTX, but I don't think the two are similar. Because CZ had no reason to believe that selling FTT would directly lead to FTX's bankruptcy. If we look back at the Bitcoin hack incident, Binance and Bitget actually lent Ether to help Bybit cover its losses and maintain its operations. Therefore, their behavior in that incident was completely different from how they are treating Hyperliquid now. I currently don't have a good theory to explain why Binance and OKX would adopt such a strategy.

Hyperliquid's behavior implicitly sends a signal that HLP has some protective mechanism. If HLP's losses are too large, Hyperliquid will intervene to protect it. From the market's reaction, Hyperliquid's price is very sensitive to changes in HLP's situation, which confuses me. I am curious about the relationship between the value of HLP and Hyperliquid. Perhaps I am missing some key points about the HLP economic model.

Tarun:

In fact, HLP does not have a clear economic model. It is more like a pure liquidity provider, not closely related to Hyperliquid's core mechanism. But I find it strange that I tend to view the HLP pool as a kind of debt instrument because it operates by raising funds from depositors.

Haseeb:

I think it is more like equity rather than debt.

Tarun:

No, HYPE is the equity. That is the confusing part.

Haseeb:

What I mean is, from the perspective of market operations, HLP investors actually receive all the profits. So it is more like equity rather than debt.

Robert:

To some extent, that is true. Market makers use the USDC deposited by users to trade across different markets.

Haseeb:

And users ultimately receive all the returns, so it doesn't resemble traditional debt.

Tom:

I believe the main reason for the decline in HYPE token prices is that this incident has created uncertainty about the future of trading platforms. After all, if a trading platform has privileged liquidity providers that never bear losses, why would anyone else trade on this platform? This is actually a problem that all trading platforms with internal market makers will face: how significant are these privileges?

Tarun:

A rather pessimistic view is that most of HLP's liquidity actually comes from the Hyperliquid team itself. So they are unwilling to bear this part of the loss.

Another reason to view HLP as a debt instrument is that it raises funds from depositors and uses these funds for market-making activities across various markets. In a sense, it plays the role of a "local lender." Similarly, protocols like Jupiter and GLP are explicitly lending protocols, charging fees in this manner. HLP profits through fees and spreads. If HLP defaults, as in this case, depositors would have priority claims.

So I think HLP is more like a debt holder, while HYPE is the real equity instrument. Because HYPE is the core asset that controls key mechanisms like the oracle, and this control is the essence of equity.

FTX Moment?

Haseeb:

Hyperliquid is more like a trading platform, while HLP is a tool used by the trading platform for market operations. HLP can be viewed as the equity of market operations, while HYPE is the equity of the trading platform itself.

Robert:

I actually think we should learn from the FTX incident, trading platforms and entities similar to hedge funds (like those teams that conduct proprietary market-making trades within trading platforms) should be completely separated. This is the only way to avoid conflicts of interest, right?

Tarun:

It is worth mentioning that Hyperliquid's mechanism is very different from FTX. On Hyperliquid, I can view every transaction of HYPE and HLP at any time and withdraw funds at any time. This transparency makes it easier to supervise. I agree with this point; I do not think Hyperliquid's approach is fundamentally wrong.

Haseeb:

If a trading platform provides protection to a market maker, clearly stating that this market maker will not incur losses, then the trading platform and the market maker are effectively bound together. It is as if the trading platform team itself is operating this market maker. If you do not believe this team can operate the market maker well, then do not invest in the equity of this market maker.

Tom:

But the question is, how is this situation presented? For example, "We have a funding pool to launch, and here is another market maker to invest in." On the surface, this seems like an open choice, but in reality, it is just a unique market maker.

Tarun:

This is indeed a unique situation. If you look at other market makers, like Seafood, he is always losing money. I do not understand why people are still willing to give him funds. His past record shows significant losses.

His funding pool is indeed interesting. But my point is that there is already a problem of adverse selection in these funding pools. It was not until this incident occurred that people truly realized how closely HLP and Hyperliquid were linked, and now this connection has become even more apparent.

Robert:

I believe they were not separated before this incident. The Hyperliquid team was operating a major market maker, and this market maker was the core market maker on the platform. Although the economic benefits belong to the users, the owners of the trading platform were actually operating this major market maker while also controlling the liquidation mechanism.

Haseeb:

Indeed, the connection between Hyperliquid and the liquidation mechanism does put them in a certain privileged position. But on the other hand, it also forces them to take on high-risk positions that other market makers may be unwilling to bear.

Robert:

Similar to Alameda, whether they want to or not, they have to bear all the bad positions on FTX, including some high-risk assets. This ultimately led to the collapse of the trading platform. Although they were forced to bear these risks, this is also a form of responsibility to some extent.

Haseeb:

Theoretically, the rationale for this arrangement is that even if HLP's funds are liquidated to zero, Hyperliquid can continue to operate. This is the ideal design. If all the mechanisms are mixed together, then the design of the entire system becomes unreasonable.

Tarun:

Sensibly, I feel that being taken down by MobileCoin feels better than being taken down by JellyJelly. MobileCoin at least tried to be a real project, while JellyJelly feels more like a joke for venture capitalists.

Haseeb:

After this incident, people might think that HLP and Hyperliquid are closely connected. This may lead to a reduction in activity from third-party market makers or liquidity providers on Hyperliquid, as they realize they are not on the same competitive level as HLP.

Tarun:

To be fair, I have observed a decrease in participation from many market makers. This phenomenon is not new, but now they have clearer reasons to reduce their activities.

Haseeb:

On the other hand, you might see more capital flowing into HLP because people now realize that the protocol might protect their investments.

Tarun:

We can use DeFi AMA as a benchmark.

Haseeb:

Will it rise or fall? So far, it has already fallen.

Tarun:

Yesterday's deposit amount was $1.85 million, three days ago it was $2.96 million, and on March 24 it was $3 million. I think this is a good benchmark time. Now it has dropped to $1.85 million. I think there are two possibilities. One is as Haseeb mentioned, that it is seen as an insurance-like product, so funds will flow in; the other is that due to a decline in confidence, trading platform fees may decrease. I am not sure which one will dominate.

Robert:

I think the risk has increased. If the incident is serious enough, the platform may intervene and close the market, setting a resolution price so that HLP does not incur losses. We just saw this situation in the JellyJelly incident. This is indeed a protective measure, but it also exposes the vulnerability of Hyperliquid's mechanisms on small assets. The likelihood of this kind of attack happening again has at least increased by an order of magnitude.

Haseeb:

I completely disagree. But now no one will try this again.

Tom:

Of course, the strategies of HLP in the trading platform are clearly also changing to reduce risk. So, this does not mean that these events are completely independent.

Robert:

But this is not an isolated incident. Two weeks ago, a similar attack occurred in the Bitcoin market, which was a very large asset. The exact parameters of the attack had already occurred two weeks ago.

The Future of 23andMe and SEI

Haseeb:

Let's talk about DeSci. Recently, there has been a major news item in the DeSci field, the SEI protocol, which is a high-performance Layer 1 EVM chain, announced that they are making the boldest DeFi investment to date.

The SEI Foundation plans to acquire 23andMe, a genetic company that recently filed for bankruptcy. They promise to protect the genetic privacy of 15 million Americans and ensure that this data is safe for future generations. They plan to migrate 23andMe's data to the SEI chain, returning data ownership to users through blockchain encryption technology, allowing users to decide how to monetize their data and share in the profits. This is not just about saving a company, but about building a future where users still have control over their most private data.

Tarun:

Does anyone on the SEI team really understand privacy protection technology? I am skeptical. If you told me that there were experts in privacy protection on this team, I would find that more meaningful. But currently, it seems like this is just a team wanting to spend a lot of blockchain funds, and their approach is even worse than those companies that are purely for acquisition.

Haseeb:

If they can achieve this goal correctly, would you support it? Maybe you could become their advisor and provide some professional advice.

Tarun:

If they are going to do this, most of the other bidders are computational biology companies, like AI drug development companies. Their goal is to use 23andMe's data to train AI models. The controversy surrounding this matter arises because many users are concerned about their data being misused. For example, eight years ago, I purchased 23andMe's service, and at that time, their privacy policy promised not to share data with third parties, but now the company is bankrupt, and this data may be used to develop drugs, which I did not consent to. This concern is understandable. So the core issues here are twofold: one is privacy protection, and the other is the method of data monetization.

What people mainly care about are the terms of service or privacy aspects, and the monetization aspect. One of the goals of all those trying to acquire the company is pure monetization, like these computational biology drug discovery companies. Then there are some non-profit organizations trying to bid, and of course, there are participants from the DeFi world.

If blockchain can really bring breakthroughs in data monetization, it could be like the ICO boom of 2017, but I doubt it will fail again. If they can truly find a way to protect privacy while achieving data monetization, that would be worth looking forward to. But as it stands, merely claiming "to let users own their data" is not enough, because so far, I have not seen any successful examples. This reminds me of Tom's earlier complaint that people complain that studios have not monetized content through blockchain, but that is not the core of the problem.

Tom:

Indeed. And I am very curious about how they will complete the acquisition. As far as I know, 23andMe has $200 million in debt. Unless they design a very complex financing structure or attract investors with SEI tokens.

Tarun:

The problem is that most of the other bidders are large companies, and SEI seems to have low odds of winning. However, emotionally, many people hope the company can find a better home rather than being acquired by bidders focused on data monetization. If SEI can propose a plan that retains the original terms of service and protects privacy, then I think they can give it a try. However, this also means they need to rely on the support of validators because they are essentially borrowing future earnings from the validators.

Robert:

From a macro perspective, all this data is currently stored in the database of a bankrupt company. Generally speaking, migrating data to the blockchain does not make it more secure than existing methods. In fact, it may increase the risk of data breaches. Of course, this depends on the security measures taken, such as encryption technology, zero-knowledge proofs (ZK), etc. But overall, I do not believe this will significantly improve privacy or security.

Haseeb:

Assuming, as Tarun said, that the current situation is that a company acquires this database and then handles this data arbitrarily. This situation is clearly not what we want to see, but theoretically, this risk does exist. I have always been skeptical about the concept of "data ownership." For example, some people propose to encrypt data and put it on the blockchain, authorizing others to use the data through decryption keys. But I have never seen this method truly solve the problem.

Tarun:

That is exactly what I am worried about. If blockchain practitioners lack understanding of privacy protection technology and attempt to handle these issues, it often backfires. For example, they may spend all their funds but accidentally leak data due to operational errors. Worse, this data might be used by certain countries to develop biological weapons.

I prefer teams focused on underlying cryptographic technology to do this, such as those researching zero-knowledge proofs or homomorphic encryption. But even these teams may not be good at commercializing technology on a large scale.

Haseeb:

It has been three months since you declared war on DeSci. How is the progress?

Tarun:

To be honest, I have given up. For example, Bio Protocol is now almost silent. I feel that everyone realizes that most of these projects are scams.

My point is that the DeSci craze is essentially just better-packaged meme coins. It has changed its branding to attract those who are tired of traditional meme coins. These people are still speculative trend-chasers. The way DeSci operates is more like donating to a non-profit organization but lacks mechanisms to verify its public welfare.

Stablecoin Legislation: Genius Act vs. Stable Act

Haseeb:

Congress is currently advancing a new stablecoin bill. Before this, we mentioned the Genius Act proposed by Kirsten Gillibrand, and now there is another bill introduced in the House called the Stable Act, proposed by French Hill.

Robert:

These two names sound like a combination of "stable" and "genius," a bit like "stable genius," don't you think? Perhaps that is where the inspiration comes from, as if quoting Trump's "stable genius" remarks.

Haseeb:

The new bill is called the Stable Act. To clarify the main differences between the Genius Act and the Stable Act: the Genius Act is more friendly to industry development. It allows banks and non-bank institutions to issue stablecoins, and state regulators can also participate in regulation, not just federal agencies. It also supports interoperability and allows for the payment of yields in certain cases, overall encouraging innovation and growth in stablecoins.

In contrast, the Stable Act is stricter. It stipulates that only banks or approved bank subsidiaries can issue stablecoins and must be directly regulated by the Federal Reserve. Additionally, it imposes more restrictions on the types of reserve assets, prohibits the payment of yields, and implements a two-year ban on algorithmic stablecoins, although existing stablecoins will have a transition period.

Robert, you have recently participated in some lobbying activities regarding stablecoin legislation in Washington, DC. How do you feel about the acceptance of this bill?

Robert:

I happened to be in Washington, DC on Tuesday and Wednesday, meeting with about 15 members of the House. Clearly, the most pressing topic on everyone's mind is stablecoin legislation.

Robert:

I feel that both sides have shown great interest in formulating legislation that is favorable to the crypto industry, and there is not much controversy. Stablecoin legislation is currently the most urgent issue because both sides recognize the need to establish a legal framework for the operation of stablecoins. This legislation is relatively straightforward, so it may become the first major crypto legislation. Although there are still some differences between the House and the Senate that need to be resolved, overall, I believe these differences will not become obstacles. After stablecoin legislation, there may be more discussions about market structure, but the current focus is still on stablecoins themselves.

At the same time, there are many discussions about the market changes that may arise after the passage of stablecoin legislation. Overall, the industry generally believes that this legislation will lay the groundwork for the future market structure. Although this goal may take time, the current discussions and focus are almost entirely centered on stablecoin legislation itself.

It is worth noting that when I communicated with some pro-crypto lawmakers, I sensed their high recognition of stablecoin legislation. Although this may carry some perspective bias, overall, the differences between the House and the Senate are expected to be smoothly reconciled, and the prospects for advancing legislation are very optimistic.

Will Stablecoins Become the "Trojan Horse" of the Crypto World?

Tarun:

I first heard about the concept of "Narrow Bank" in 2009. At that time, many people were discussing narrow bank legislation. Although this makes me sound a bit old-fashioned, back then, everyone was exploring this model: should there be a strictly limited bank that can only provide very basic types of yields?

Haseeb:

Can you explain what a narrow bank is?

Tarun:

The definition of a narrow bank has evolved over time, but the core idea is to simplify banking. Especially after the financial crisis, some proposed whether banks should be subject to stricter regulations, such as limiting their participation in trading or other complex activities? Or could a bank be created that only provides basic services, like deposits and loans, without engaging in other complex businesses? Interestingly, many early fintech applications, to some extent, resembled "pseudo-narrow banks." They allowed users to deposit funds but offered almost no yield products. Users might indirectly purchase government bonds through these platforms or provide Bitcoin services like Square, but these platforms themselves do not engage in complex investment activities like proprietary trading or bond portfolio investments.

In a sense, many of the stablecoin bills remind me of the concept of narrow banks. Stablecoins themselves do not yield, and the way the banking license is used behind them seems very interesting to me. This idea of narrow banks has been around for nearly 20 years since it was first proposed, and now it is finally being realized through blockchain technology. I feel like history is repeating itself, just at a very slow pace. After all, the U.S. went nearly a decade without new banks being established or new bank licenses being issued.

Robert:

My understanding is that the core of a narrow bank is to keep all deposits at the Federal Reserve's discount window, maintaining 100% liquidity. This way, the bank does not need investment analysts or loan officers; all deposits are handed over to the Federal Reserve to earn interest, and then a certain fee is deducted and paid to depositors. In a sense, this is basically a branch of the Federal Reserve.

This model is a full-reserve bank, with 100% liquidity, without worrying about liquidity risk. Theoretically, it could manage a large banking system with just a dozen employees. But the reason people oppose narrow banks is that they would compete with existing commercial banks. Commercial banks expand the money supply through loans, while narrow banks simply store funds at the Federal Reserve, which would reduce the liquidity of quality assets like mortgages.

Haseeb:

I think the reason the Federal Reserve rejects narrow banks is that it undermines its ability to intervene directly in the money supply. While mortgages can still be provided by private lenders, once the market fully shifts to private lending, the Federal Reserve will lose direct control over the expansion of the money supply.

Robert:

From another perspective, the Federal Reserve may actually gain more intervention capability because overnight rate adjustments will affect all market participants.

Haseeb:

If there is still a reserve ratio, then yes. The reserve ratio is the second lever, which is a very powerful lever that can instantly change the money supply. Raising or lowering interest rates is clearly subject to a zero lower bound, although technically you can break through it, but the U.S. will not enter negative interest rates. But this is just a slower mechanism; entering the market, as a bank, you can use everything in reserves for investment, and this change will be faster.

Robert:

This reminds me of the Genius Act. Tarun said stablecoins are like narrow banks, but I think they are not entirely the same.

Haseeb:

I think he is referring to the Stable Act, especially because it does not allow yields. Why are yields not allowed in the Stable Act? Probably because they do not want it to compete with commercial banks.

Robert:

Tarun's point might be that this restriction makes stablecoins more like narrow banks. But the core of a narrow bank is that it allows for the payment of fully liquid interest.

Haseeb:

So, if yields are not prohibited, you could create a narrow bank with stablecoins. Therefore, under the Stable Act, you cannot really create a narrow bank that competes with commercial banks. But under the Genius Act, you could essentially have a stablecoin that only holds government bonds and returns all the government bond yields minus 20 basis points or other yields, which would ultimately become a very simple business model.

You could say this is Tether's model; clearly, they do not pay yields, but if they did, that would be an incredible business model. Its labor efficiency is very high; they have about 90 employees managing over $100 billion in assets. So this is quite a good business.

Haseeb:

I think this point makes sense. Stablecoins may reintroduce the concept of narrow banks in a more acceptable way while also bringing geopolitical advantages. In contrast, narrow banks would only impact commercial banks without helping the internationalization of the dollar. The advantage of stablecoins is that even if they compete with commercial banks, they can expand the overall market size of the dollar through internationalization. Narrow banks cannot achieve this because they are in a zero-sum game between commercial banks and narrow banks.

From a policy perspective, this is why stablecoins may be more favored. But I also agree with your point, Tarun, that when central bankers or bank executives examine this issue, they may lean towards monopolizing the issuance of stablecoins to bank license holders. This is actually a manifestation of "regulatory capture," protecting existing interests by restricting market participants.

Robert:

What is your view on the final bill? Do you think it will be more like the Genius Act or more like the Stable Act, with fewer restrictions or stricter ones?

Robert:

I think, aside from the yield aspect, it will be less strict. I believe the current commercial banking sector does not want to see yields on stablecoins.

Haseeb:

This reminds me of some strange phenomena in the banking industry. For example, my Chase bank account, why can't my cash automatically transfer to a money market account to earn yields, and I have to rely on my active operation? It would be great if banks could automate these operations. But in reality, many people do not operate actively, leading to cash sitting idle. This phenomenon is actually very common; although users can transfer funds to a money market account with the push of a button, many people just do not click it. Therefore, brokers make a lot of money from this laziness.

Robert:

One of the main sources of income for brokers is the interest spread.

Tom:

I heard that the FTC investigated Citibank because they launched two almost identical savings products, but one had a lower interest rate. This indicates that banks profit from information asymmetry, while stablecoins avoid this issue in some aspects.

Robert:

You cannot easily lower the interest rates for new customers, but you can launch a second product while keeping all old customers at the same stagnant rates.

Haseeb:

But ironically, if you view this as a cash account, stablecoins, even if you do not earn yields, for example, not earning yields on Tether or USDC, simply borrowing in the market can still yield quite high returns.

Tom:

Currently, the borrowing yield in the market is between 5% and 6%. The advantage of narrow banks is that users can freely choose their risk allocation rather than have banks make decisions for them. For example, they can choose to invest in private credit or tokenized government bonds instead of being bundled together by banks. If users are willing, they can operate themselves.

Haseeb:

This makes a lot of sense. If stablecoins really draw deposits out of the banking system, I guess this might be because they allow users' cash to generate returns, no matter how lazy they are.