Overnight, users in the crypto industry suddenly found they could buy on-chain stocks of Nvidia and Tesla using their stablecoins, as if they had fallen down a rabbit hole into a wonderful alternate world.
1. This Episode Introduction
On July 31, SEC Chairman Paul S. Atkins delivered a speech titled "America's Leadership in the Digital Financial Revolution" in Washington, D.C., unveiling a plan called "Project Crypto"—this reform plan aims to fully transition the U.S. financial markets on-chain.
Atkins stated that he would provide regulatory exemptions for "the issuance of tokenized securities within the United States" where appropriate, to ensure that the American public is not excluded. "Broker-dealers with ATS should be allowed to offer non-securities crypto assets, crypto asset securities, traditional securities, and other services on their platforms," allowing market participants to create "Super-Apps." To support on-chain trading of tokenized securities, he also suggested that "Reg NMS may need to be revised to correct the distortions it has caused in the current market."
This podcast episode was actually recorded before Atkins' speech. However, even without the formal launch of this plan, the on-chain transition of the U.S. financial market, represented by U.S. stocks on-chain, has already begun with great momentum. From Dinari to xStocks, from Gemini to Robinhood, overnight, users in the crypto industry suddenly found they could buy on-chain stocks of Nvidia and Tesla using their stablecoins, as if they had fallen down a rabbit hole into a wonderful alternate world.
"In the next three to five years, traditional stock exchanges and crypto exchanges may become direct competitors, such as the competition between Nasdaq and Binance, or Hong Kong Stock Exchange and Coinbase." "The significance of on-chain U.S. stocks may not just be an alternative channel, but truly becoming a financial infrastructure." "Whether it’s brokerages, asset management, or exchanges, fundamentally, what users rely on is trust + local experience. It’s hard for the market to form a super unified platform; there will always be second, third, and fourth players who can capture their respective shares and thrive."
These exciting insights come from the sixth episode of "支无不言," featuring three guests from frontier technology investment, internet brokerages, and on-chain stock startups. Now, let’s dive in together and explore the vast ocean of everything on-chain.
Guest Didier/Zheng Di
Frontier technology investor, managing the knowledge community "Dots Institutional Investor Community."
Guest Sherry Zhu
Global Head of Digital Assets at Futu Group, previously worked at the Hong Kong Securities and Futures Commission for over five years, responsible for crypto policy and licensing.
Guest Zixi Zhu
Founder of the on-chain stock project Stable Stocks, former employee at Matrix Partners, co-founder of 10K Ventures, X: Zixi.sol.
Host Hazel Hu
Host of the podcast "支无不言," with over 6 years of experience as a financial media reporter, core contributor to the Chinese Public Goods Fund GCC, focusing on the practical applications of crypto. X: 0xHY2049; Jike: A heartless Yuyue.
Host Ivy Zeng
Host of the podcast "支无不言," previously worked in VC post-investment, involved in pop-up city and payment connections, currently responsible for growth at a new type of bank. X: IvyLeanIn; Jike: Spoon in the cup; Xlog: ivyheretochill.
Sponsor
This episode is sponsored by Cobo, the largest digital asset custody and wallet solution provider in the Asia-Pacific region.
In the face of the global payment wave, Cobo helps enterprises build digital currency acquiring capabilities, providing a full-stack payment solution covering underlying wallets and risk control compliance.
Visit https://www.cobo.com/ for more information or click to read the original text to schedule a demo.
2. User Demand and Profile for On-Chain Stock Trading
Ivy: Before we get into the main topic, let’s have a little interaction: Do any of the guests currently have U.S. stock accounts? Which software are you using? Are you considering investing in on-chain U.S. stocks?
Sherry: I am a deep user of Futu. I initially chose Futu because of its excellent product experience, and later joined the company because I was impressed by its product capabilities. I have always been very interested in innovative products like on-chain U.S. stocks.
Currently, the U.S. stock tokens launched are mostly in derivative form, which aligns more with the habits of crypto-native users rather than traditional U.S. stock investors who focus on voting rights, dividends, and other entitlements. However, these on-chain derivatives do bring broader application scenarios, such as serving as underlying assets for DeFi Lego. As long as this product addresses real user pain points, I believe the market will respond positively.
Zheng Di: I initially used IB (Interactive Brokers), but recently I have also started using Futu's overseas version, Moomoo. The interface design of Chinese companies is more in line with the habits of Chinese users.
I believe that in this wave of market trends, the real beneficiaries are internet brokerages. Traditional exchanges (CEX) face greater challenges because most of the crypto market speculation has resulted in significant losses, while users trading U.S. stocks have made substantial profits, prompting many CEX to start operating accounts on Xiaohongshu for traffic.
Ivy: So, Mr. Di, are you considering trading on-chain U.S. stocks now or in the future?
Zheng Di: I think it’s possible. Because relatively speaking, on-chain trading is indeed more convenient.
First, the money in the crypto space and the money in traditional brokerage accounts are two separate systems that generally do not interconnect, which is like fighting on two battlefields. The "friction" in the deposit and withdrawal process is quite severe, especially with fiat currency. Even in regions like Singapore or the U.S., where deposits and withdrawals are relatively easy, the costs are not low.
For example, if you deposit through OTC (over-the-counter trading), you have to bear fees of several thousand; if you use a licensed exchange like Coinbase in Singapore, you also have to add about 1% in fees and 9% in consumption tax. Once the bill comes out, you really feel like you might "faint."
So basically, everyone still tends to manage these two parts of funds separately and is not very willing to frequently transfer across systems. If on-chain trading can provide sufficiently deep liquidity for U.S. stocks, it would definitely be worth considering.
Zixi: I mainly use Tiger Brokers because when I was studying, Futu had not yet launched in Singapore. Tiger supports PayNow deposits, which is very convenient. I have also used RockFlow, but its deposit and withdrawal speed was too slow.
Ivy: Today we are discussing on-chain U.S. stocks, and the key question is, who really needs this product? Zixi, before you decided to enter this field and start your business, you must have done more market research than we did. Can you share your insights on this?
Zixi: Of course. I categorize users of on-chain U.S. stocks into three types:
Novice Users: These users are mainly located in countries with strict foreign exchange controls, such as China, Indonesia, Vietnam, the Philippines, and Nigeria. They have stablecoins but cannot open bank accounts overseas due to various restrictions, making it difficult to buy traditional U.S. stocks. On-chain U.S. stocks can bypass complex KYC and deposit/withdrawal processes, allowing them to easily invest in American stocks.
Professional Users (Pro Users): They have both stablecoins and overseas bank accounts, and may have accounts with Futu or Tiger. However, their pain point is that traditional brokerages offer too low leverage, for example, Tiger's leverage is only 2.5 times. On-chain, by setting a higher LTV (loan-to-value ratio), high leverage can be achieved; for instance, with an LTV of 90%, one can achieve 9 times leverage trading.
High Net Worth Users (Pro Max Users): These individuals hold U.S. stock assets long-term and may earn interest, dividends, or enjoy the benefits of stock price increases through margin trading in traditional brokerage accounts. Here, we can tokenize their stocks, for example, turning them into sTSLA or sNVDA (stablecoin versions of Tesla or Nvidia), and then these tokens can be used for LP, lending, or even cross-chain operations on-chain.
Hazel: Thank you for sharing, Zixi. Sherry, you are closest to traditional stock users. From the perspective of internet brokerages, what user needs currently cannot be met by non-on-chain solutions?
Sherry: I can summarize a few common pain points:
Low capital efficiency: Traditional securities trading has a T+1 or even longer settlement cycle, leading to slow capital turnover, especially when investing across markets, where fund transfers take even longer.
Trading time restrictions: U.S. stocks open at night, making it difficult for Asian users to monitor the market. Many users hope for a 24/7 trading window.
High investment threshold: Stocks like Tesla and Nvidia have high prices, making it difficult for users who want to invest small amounts to enter.
On-chain stocks provide an imaginative space, such as real-time settlement, fractional investment, 24/7 trading, on-chain collateral lending, etc., which are scenarios that traditional finance struggles to cover.
Of course, Futu is also innovating to address these issues:
Launching a unified purchasing power mechanism that can calculate users' total assets in cash, securities, and money market funds across multiple markets in real-time, enabling cross-market trading;
Launching night trading, covering almost 5×24 hours;
Launching Fractional Shares, allowing users to buy Tesla stock for as little as $5.
However, I must emphasize that despite our continuous innovation, the underlying structure is still centralized. The emergence of blockchain could indeed bring about deeper structural changes, even potentially driving the reconstruction of the entire financial infrastructure.
3. Long-Term Trends of Stock Tokenization
Hazel: Mr. Di, we have heard some data, such as XStocks having a total market value of about $30 million, with daily trading volumes increasing by tens of millions of dollars. As a researcher and investor, how do you view the data on on-chain U.S. stocks?
Zheng Di: Currently, the activity of on-chain U.S. stocks is still concentrated on a few centralized exchanges, such as Gate and Huobi, which account for 70%-80% of the trading volume.
However, products like those in cooperation with Kraken have not performed prominently on-chain, despite their official announcements of many news items, such as holding licenses in Bermuda through subsidiaries; overall trading volume has not picked up.
From a compliance perspective, if we were to rank them, Robinhood is the most compliant, Dinari is in the middle, and Kraken is taking a more aggressive approach.
An important regulatory trend is that SEC Commissioner Hester Peirce recently stated: if stock token trading is to be opened to retail investors, it must be conducted on platforms that have obtained national securities exchange licenses. This means that only platforms like NYSE or Nasdaq can do so; otherwise, it can only be limited to qualified investors. This statement may have a significant impact on projects like Dinari and Coinbase.
But in the long run, I believe the integration of stock tokens is a trend. In the next three to five years, traditional stock exchanges and crypto exchanges may become direct competitors, such as the competition between Nasdaq and Binance, or Hong Kong Stock Exchange and Coinbase. By then, traditional exchanges may also move their trading architecture on-chain, significantly reducing back-office settlement costs.
So we need to consider whether the role on-chain will become a front-end tool directly facing users, or whether it will become the back-end infrastructure for brokerages and exchanges. I think both possibilities exist.
One scenario is that end users trade directly using on-chain products; another scenario is that users still place orders through familiar apps, brokerages, or front-end operating systems, but the entire underlying trading logic is already running on-chain. The trend behind this is irreversible.
Take Robinhood as an example. Although it currently uses contracts for difference (CFD) to "lightweight and quickly comply" with stock token trading, its business model inherently favors an on-chain structure. So I predict that Robinhood's first phase is CFD, and in the second phase, it will definitely move towards a full on-chain trading process.
Currently, trading of on-chain stock tokens is still mainly conducted on centralized exchanges (CEX), and truly full-chain scenarios are still relatively rare. However, there is another possibility: users may be using CEX or brokerage front-ends, but the back-end trading and settlement architecture is actually running on-chain. This "superficially centralized, but back-end on-chain" model may be the most suitable compromise that aligns with current user habits while accommodating new technological architectures.
Especially in the short term, the on-chain threshold remains high, and novice users are unlikely to interact directly with the chain. Perhaps in the future, when AI agents are mature and security solutions are resolved, users will be able to trade on-chain assets through AI self-service, but that will take a few more years. So before that, allowing users to indirectly access on-chain assets through user-friendly CEX or brokerages may be the mainstream form. I believe we are still in an early stage, but this path is certain and has the potential to explode.
4. Customer Acquisition for On-Chain U.S. Stocks: A Hybrid of Web2 and Web3 Marketing?
Hazel: There’s an extended question about user demand—how to attract these users? After all, the product of on-chain U.S. stocks itself spans both Web2 and Web3, so will its marketing also be like a "Frankenstein"? Zixi, as a startup, what is your GTM (go-to-market) strategy?
Zixi: From day one, our product has been a broker. Retail investors deposit USDT into our wallet address, and when placing an order, we transfer the funds to a U.S. bank account via Coinbase, then into Nasdaq to complete the transaction. After the transaction, we mint a 1:1 stock token for the user on-chain based on the information returned by the brokerage. Throughout the process, the information flow and asset flow are separated; we rely on the information flow to ensure the on-chain mapping is trustworthy.
Our methods for attracting users are also divided into two levels:
On-chain level: We have a product called vote, where users stake tokens like sTSLA and sNVDA into vote. We then provide these tokens to different DeFi protocols, such as lending and DEX market-making. We share profits with market makers, returning most of the earnings (80-90%) to users.
Operational level: We mainly rely on Web3 strategies, such as trading mining and liquidity mining. These growth methods are more native and can better stimulate the enthusiasm of crypto users.
Sherry: I think the two directions Zixi mentioned are great, and I can add another direction—the demand for institutional arbitrage.
For example, market makers and quantitative funds are also interested in on-chain U.S. stocks and can enter through arbitrage mechanisms. However, institutions are more sensitive to regulatory frameworks and must meet internal compliance requirements to participate.
So I believe the growth path looks like this:
Initially relying on the Web3 native ecosystem to feed back users, attracting capital flow through on-chain strategies;
In the medium term, rapidly iterating products in the gray area where regulations are not yet clear;
In the long term, forming a Matthew effect that redefines the way capital flows.
Of course, you could say there are elements of "regulatory arbitrage" in between, but perhaps it can force reforms, just like the wild growth of USDT ultimately led to regulatory and financial structural changes. The significance of on-chain U.S. stocks may not just be a "substitute channel," but truly becoming a financial infrastructure.
5. Global Expansion of Chinese Enterprises
Ivy: Zixi, when we heard you talk about this direction a month ago, on-chain U.S. stocks were still a nascent market. But in the past half month, various players have suddenly "moved," with new features launched by internet brokerages and exchanges following suit, rapidly increasing market enthusiasm. As a professional startup, how do you view the competition? What potential differentiation opportunities exist?
Zixi: I think players in the on-chain U.S. stock market can be roughly divided into three categories:
U.S. project teams, such as Dinari and Robinhood. Dinari has a better compliance foundation and mainly focuses on the domestic U.S. market; while Robinhood, due to compliance restrictions, is currently only testing the waters in the European market, with a weak presence in the Asia-Pacific region.
European project teams, such as Backed Finance. They generally collaborate with exchanges, taking a B2B2C approach, mainly focusing on the European market. For Asian markets, especially C-end users in the Asia-Pacific region, the reach is quite distant.
Our Chinese-speaking startup team has been more familiar with regions like Asia-Pacific, Southeast Asia, and the Middle East from the beginning. Many members of our team have lived or worked in these areas, so we naturally understand local users, languages, cultures, and compliance pathways better.
To be honest, entering the U.S. and European markets is not easy for us either. But we are very familiar with the local market, which gives us a significant advantage in onboarding institutional users and communicating with retail investors. For example, many users may ask why, as a Chinese user, you don’t use Interactive Brokers but choose Futu or Tiger? These local brands serve as trust anchors.
So you will find that whether it’s brokerages, asset management, or exchanges, fundamentally, what users rely on is "trust + local experience." It’s hard for the market to form a "super unified" platform; there will always be second, third, and fourth players who can capture their respective shares and thrive.
Ivy: It sounds like the on-chain U.S. stock market still has a non-oligopolistic structure, with each player having their own advantages. I would like to ask Sherry if you could introduce Futu's global business layout? Especially in the Web3 field, what capabilities does Futu currently possess?
Sherry: Sure, Futu has indeed made a lot of efforts in globalization over the past few years. We use the "moomoo" brand in markets outside of Hong Kong. Since entering the U.S. market in 2018, we have gradually expanded our business to Singapore, Australia, Japan, Canada, and Malaysia, and recently entered the New Zealand market through an Australian license. Currently, we have over 26 million registered users globally, with client assets reaching the hundred billion dollar level, and annual trading volume approaching one trillion dollars.
In the Web3 direction, we actually laid the groundwork early, mainly divided into the following areas:
Hong Kong: Starting in August 2023, Futu began supporting trading of mainstream cryptocurrencies (such as BTC and ETH). This year, we also launched deposit and withdrawal functions, allowing users to convert crypto in their wallets to fiat with one click, directly investing in Hong Kong stocks, U.S. stocks, and other traditional assets, or withdrawing to banks. This greatly resolves the issues of "difficulty in withdrawing funds, high fees, and non-compliant channels" for crypto users.
Additionally, Futu has set up a licensed exchange called "Cheetah Trading" in Hong Kong, which obtained a Virtual Asset Trading Platform License (VATP) from the Hong Kong Securities and Futures Commission in January this year and is currently cooperating with regulators to complete subsequent evaluations.
Singapore: moomoo is the first local digital brokerage to obtain a digital asset trading license from the Monetary Authority of Singapore. Crypto trading was launched in July last year, and this year we will also launch deposit and withdrawal services, with plans to support more cryptocurrencies in the future.
United States: We have obtained MTL licenses and exemptions in over 40 states, providing trading for more than 30 mainstream cryptocurrencies.
In the future, we hope to expand our Web3 business to all countries covered by securities business under a compliant framework, achieving the vision of "one app, one account, investing in global assets."
Hazel: Recently, certain virtual asset concept stocks in the Hong Kong market surged. What do you think about this market enthusiasm?
Sherry: Indeed, this indicates that the market holds an optimistic view of traditional financial institutions entering Web3, which is a typical sign of a bull market. However, it also needs to be viewed calmly:
Some brokerages have actually obtained upgraded qualifications based on a primary license, essentially still acting as brokerages, providing virtual asset services through comprehensive accounts. The Hong Kong Securities and Futures Commission's official website has currently disclosed that 43 institutions hold similar qualifications. Under this model, brokerages can only be bound to licensed virtual asset exchanges, and their business scope and tradable cryptocurrencies are restricted by upstream exchanges.
We at Futu hold both a primary license and an exchange license, aiming to create an "integrated upstream and downstream" ecosystem, generating synergistic effects in stablecoin applications, asset issuance, and other areas.
But it should be emphasized that a license is a ticket to entry, not a guarantee of success. In the long run, those platforms that can organically integrate both types of resources, relying on a strong customer base to form ecological effects, and possess continuous innovation capabilities will truly be competitive.
Hazel: Zixi, what compliance licenses are involved in your current business operations at Stable Stocks? What is your strategy?
Zixi: Currently, our team does not hold licenses directly, but we collaborate with partners through license authorization, specifically divided into two parts:
1) Broker Dealer Licenses
This area is particularly focused on by the SEC. We conduct actual stock trading off-chain, which falls under tokenized securities, so we must operate based on institutions with securities licenses. The licenses we use include:
Hong Kong's primary license;
U.S. Broker Dealer license;
Australia's AFSL (Australian Financial Services License), etc.
These are held by our partners, who allow us to operate together within a compliant framework.
2) MSB Payment Service Licenses
Since we also involve the process of USDT or fiat entering the banking system from Coinbase and then transferring to the exchange's bank account, which involves compliance for deposits and withdrawals, we also need:
U.S. Money Service Business (MSB) licenses;
Similar payment permits in other countries to legally explain the source and flow of user funds.
In addition, if we launch asset management products in Hong Kong in the future, we may also need to apply for a Type 9 license; if we expand in the Middle East, such as Dubai and Abu Dhabi, we will also need to obtain virtual asset operating licenses like VARA and ADGM.
6. The Ambitions of Robinhood and U.S. Brokerages
Hazel: We just talked about Chinese brokerages and crypto-native startups, but there is another important player—the internet brokerages in the U.S. stock market. Among them, Robinhood is a very representative company. Di, I know you are very familiar with Robinhood's business model and history. Can you help us break down their development path, revenue composition, and why they are moving towards on-chain U.S. stocks or CFDs?
Zheng Di: Robinhood is actually a very "divided" company. On one hand, its zero-commission model and young user base are enviable; on the other hand, many people feel that its model is somewhat "exploitive." It is distinctly different from traditional internet brokerages like Interactive Brokers (IBKR).
The revenue structure of Interactive Brokers is relatively stable, mainly coming from:
Interest income (including interest on idle cash from clients and securities lending interest), which can account for 60-70%;
The remaining 30-40% comes from trading commissions and other fees.
In contrast, Robinhood's revenue mainly comes from:
PFOF (Payment for Order Flow) order flow sale income;
And a portion of interest income (recently they have been working to increase this proportion, such as improving the penetration rate of their Gold membership, which has now reached 12.3%).
However, Robinhood cannot completely escape PFOF because its users are fundamentally different from those of Interactive Brokers:
Interactive Brokers targets institutions and high-net-worth investors who are price-sensitive and very concerned about fee structures.
Robinhood, on the other hand, targets younger retail investors who have a higher risk appetite and are less price-sensitive, focusing more on "ease of use" and "excitement."
This also determines Robinhood's PFOF revenue model—you are "charged" without realizing it.
Zheng Di: To illustrate with data, for example, when Robinhood conducts stock trading, the market-making rebate from Citadel Securities is 0.8 basis points, which is 0.008%; but for options trading, this rebate can reach 8 basis points, ten times that of stocks.
If it is crypto trading, the rebates from previous players like Binance.US, C2, and Jump are about 35 basis points, which is 45 times that of stocks and 4.5 times that of options.
And it doesn't stop there; they also have premiums and slippage from their smart routing, which can contribute another 20 basis points. So you can understand that Robinhood makes about 55 basis points of revenue on each crypto trade, which is already half of Coinbase's retail fees. But everyone thinks Coinbase's fees are high because they are openly charged, while no one thinks Robinhood is expensive because Robinhood is a brokerage, not an exchange, using a "hidden fee" model.
This is also why the Florida Attorney General investigated Robinhood a few years ago, believing it was suspected of false advertising. You claim to be "zero-commission trading," but in reality, you are charging through PFOF and hidden slippage.
A few days ago, when I was communicating with some institutions, they were also confused: why would European users use Robinhood for CFD trading of U.S. stocks instead of local brokerages?
I said it’s simple, because Robinhood claims to be zero-commission trading. There may not be as competitive zero-commission brokerages in Europe. The only apparent charge from Robinhood is a 1‰ currency exchange fee for euros to dollars; everything else is 0. But they can use other methods, for example, if Tesla's stock is $200, they might quote the user at $200.05 or $201, making an extra dollar that the user is completely unaware of. This structure determines that the real cost of Robinhood's "zero-commission trading" is not low.
So Robinhood's profit source relies on feeding prices and hidden fees. Because of this, Robinhood's business model is very dependent on the "characteristics" of its user base—young, willing to take risks, and lacking financial literacy. Of course, this model is very friendly to Robinhood's shareholders.
Hazel: I would also like to ask Di to explain further. We know that Robinhood essentially operates CFDs, but many listeners may not fully understand this model. Can you explain in simpler terms what CFDs are and how they anchor to the underlying stock prices?
Zheng Di: Sure, you can simply understand CFDs (Contracts for Difference) as a type of perpetual contract with one times leverage. It is cash-settled, and as long as you do not close the position, it continues to exist.
But this is actually an "air contract" because you cannot withdraw coins, and there is no real stock asset delivery. The reason it can operate is that you need an "anchor"—which is the actual price of the underlying stock.
There are two ways to obtain the anchor price:
Relying on arbitrageurs to smooth out the price difference, but this can lead to uncompensated losses, which is a drawback of many DeFi models;
Introducing a price feed mechanism (Oracle), which is the background for the rise of Chainlink during the last bull market.
But Robinhood is clearly unwilling to let profits be taken by arbitrageurs; its entire market is a closed loop, and it can feed prices itself. I saw in Backed Finance's working documents that they use an oracle once every 24 hours, mostly using their own oracle, and will urgently use the oracle for intervention when the price difference exceeds 10%. To prevent arbitrage, they also set a redemption limit; if you are an institutional investor, the highest tier investor, then your total subscription and redemption amount within 24 hours seems to be capped at 30 million Swedish Krona.
Conversely, Robinhood likely has a similar mechanism. When the price deviates too much from the underlying stock, it triggers a custom price feed. But this price is set by Robinhood itself, meaning they have the final say.
So their zero-commission model makes money not from fees, but from price differences and slippage, and users are completely unaware of it.
Moreover, unlike Dinari or xtock, which do 1:1 asset mapping, Robinhood's CFDs are legal and compliant securities derivatives. They are completely compliant, and there is no issue of holding assets on behalf of others. Because we all know that during the GME incident, Robinhood once faced a $3.7 billion funding gap and nearly went bankrupt. Later, the U.S. Congress, including the SEC, also attempted to investigate Robinhood. So I believe that in the years following that incident, Robinhood has been under continuous scrutiny in the spotlight.
For example, Robinhood holds the underlying stocks in the U.S. not to collateralize the CFDs, but to hedge its own short positions—because CFDs are essentially a bet between Robinhood and the client: you are long, and they are short. So they need to hold the underlying stocks to hedge. This entire design is very clear in terms of compliance, risk control, and profit structure. It can even be said that this model is much "cleaner" than doing token mapping on-chain. But I also believe that in the future, Robinhood will definitely venture into on-chain because on-chain products are easier to profit from. The rebates for on-chain products are much higher than for stocks, providing greater profit margins.
7. Delving into the Impact of Stock Tokenization on the Profit Distribution Chain
Ivy: Just now, Di introduced how Robinhood achieves on-chain U.S. stocks and how it tracks prices. Zixi, can you also explain how Stable Stocks implements price feeds? What stakeholders are involved in this process? How do you handle prices during weekends or after market hours? What should be done when there is a price difference between on-chain and the underlying stock?
Zixi: As previously mentioned, simply put, users can buy and sell U.S. stocks through our broker. When a user buys U.S. stocks on-chain, we will synchronously purchase the actual stocks off-chain through our partner broker and mint the corresponding tokens to achieve a 1:1 mapping.
Many people ask us: Don’t you need an oracle to synchronize prices? In fact, in our model, the trading price on-chain is essentially the broker's price. The user's trading behavior directly reflects the real market price, and there is no possibility of decoupling. Theoretically, there should not be a significant price difference.
Of course, not everyone can participate in arbitrage (for example, cross-market arbitrage or "brick moving"). But as long as we can connect more broker interfaces in the future, such as being able to open accounts on Futu, Tiger, etc., and operate arbitrage during night trading, then the price differences will be further smoothed out.
After market hours or during weekends is a relatively special time period. Because the off-chain market is closed, but the on-chain market is still open.
Our current idea is to calculate a reference price for this period using a weighted method, such as:
Friday closing price × α + on-chain trading average price × β
We are currently further researching this model with the team to ensure it is both reasonable and reflects market expectations.
Ivy: You just mentioned that the old model of Mirror Protocol is quite different. Can you elaborate on how you differ from it?
Zixi: Sure. I have used both Mirror Protocol and Synthetix. Their core issue is low liquidity utilization efficiency. For example, Synthetix requires 80% over-collateralization; if you deposit 100 USDT, you might only be able to mint synthetic assets worth 20. The capital efficiency is too low.
In contrast, we completely map the real purchasing behavior of off-chain stocks 1:1. As long as you buy off-chain, we mint the corresponding tokens. This is similar to the logic of stablecoins, with higher liquidity efficiency. We hope to enable on-chain users to access quality assets, achieving true financial inclusivity.
Hazel: So specifically, how does Stable Stocks make money? How is the profit distribution structured among different stakeholders like brokerages, public chains, and exchanges?
Zixi: Our current profit model can be divided into three parts:
1) Broker Transaction Fees (Transaction Fee)
We operate as a proprietary broker, similar to platforms like Futu. As long as we do not adopt the PFOF (Payment for Order Flow) model, we must charge a "full fee," which is the standard trading commission. Since we do not make money by selling orders, this part of the fee will be higher than that of Robinhood or Futu.
Additionally, we also take on some roles in OTC services. We have an MSB license, allowing us to legally provide OTC trading, rather than seeking liquidity from the black market. However, the volume has not picked up yet, so the rates charged to VIP users are low, while ordinary users will be charged higher rates.
Furthermore, we will also collect a portion of transaction fees from on-chain trading, expected to be between 0.15% to 1% (specific rates are yet to be determined). Overall, the average fee across the platform is around 0.14%.
2) On-Chain Business Model: Market Making and Liquidity Provision
On-chain, our main profit model is market making and providing liquidity. We are building a liquidity pool similar to MakerDAO, where users entrust their assets to us, and we seek professional market makers, DEXs, and lending protocols for liquidity distribution.
The logic is similar to Lido's staking model for Ethereum:
Ordinary users cannot become nodes or traders themselves,
So they entrust their assets to us,
We configure liquidity, hedge risks, and can also receive protocol subsidies.
Users receive yield certificates, and we help them split their stocks to use for different strategies in liquidity, which will become another source of income for our platform.
3) Off-Chain Cooperation and Lending Business
In the future, we will also have some income related to physical stock lending. This part mainly comes from cooperation with brokerages, such as helping clients with stock pledges and wealth management services.
Ivy: Alright, let's ask Di about the previous question. What do you think about the changes in the brokerage profit model after stocks are tokenized? Especially regarding the new profit-sharing structures that may arise between brokers, issuers, and public chain exchanges?
Zheng Di: This is a great question, and it actually needs to be viewed from two dimensions: legal definitions and business models.
From a legal perspective, no matter how you tokenize it, it is still a security. Even if it only involves the transfer of rights to income without complete shareholder rights, it will still be recognized as a security. Therefore, regulation will still manage it as a security.
However, from a business model perspective, whether the rebate for tokenized stocks is calculated based on the traditional stock rate of 0.8 basis points or the DEX rate of 35 basis points depends on the capabilities of the on-chain matching party—whether it can control slippage and profit margins.
In fact, there are already many tools on-chain to prevent "slippage" and "order eating," but the problem is that front-end novice users find it difficult to perceive these behaviors. This means that even if backend costs decrease significantly, the front-end quotes may not necessarily adjust downwards by the same amount.
Let me give you an example: this has already happened in the payment industry. Stripe acquired Bridge, which provides stablecoin payment services, and the service quotes for B-end enterprises are only 1.2‰ to 2.5%, while Stripe's traditional card organization charges 2.8% to 3%. After the acquisition, Stripe packaged the stablecoin service as a new product and quoted it at 1.5%.
This seems like a "price reduction," but the actual cost is probably only 0.12‰ to 0.25‰. Even if you add in fiat currency deposit and withdrawal costs and OTC costs, it won't exceed 0.18%. This means they are still making a lot of money, just that "the quote hasn't dropped much."
This logic also applies to Robinhood. If stocks are tokenized in the future, they may no longer rely on the underwhelming rebate business off-chain to make money, but instead use the extremely low financial transaction costs on-chain to reduce their costs while maintaining a "reasonable quote" on the surface, charging 5 basis points or 10 basis points is still much higher than what they currently charge.
Even if the on-chain rebate is only 8 basis points, that is still ten times the 0.8 basis points off-chain. So Robinhood will definitely embrace this model because it offers significant revenue growth potential.
But this also depends on whether regulation can keep up. If regulation remains absent, it becomes a gray area—intermediaries can continue to exploit information asymmetry to "harvest" novice users. This is a future that Robinhood may hope to achieve.
Hazel: I previously mentioned to you that when chatting with GPT, it believed that regulation would "penetrate," meaning that as long as it is essentially a security, the PFOF rebate model would still be regulated. But you said GPT was greatly mistaken. What do you think?
Zheng Di: I still hold this view: even if regulation takes action, it will be relatively slow. The U.S. has not completely banned market maker rebates; it only requires "transparency."
Currently, countries like the EU, the UK, Canada, Australia, and Singapore have already banned this model, but the U.S. has not. Because Robinhood is like the sales department of an investment bank, and Citadel is like the trading desk of an investment bank. Traditional investment banks internally have these two departments cooperating, but Robinhood and Citadel are "external collaborations." If rebates are really banned, then Robinhood and Citadel could simply merge into one company and still find a way around it.
In the current political structure, the Republican Party generally does not care about this issue, while the Democratic Party is more concerned. I originally thought we would have to wait until the midterm elections in November next year for the Democrats to push for regulation, but I didn't expect Florida to start investigating so quickly.
Of course, Florida cannot directly regulate PFOF; it can only approach it from the angle of false advertising—such as Robinhood claiming "zero-commission trading," but in reality, it has made a lot of money through order rebates, which can only be used as a reason for investigation.
This is worth ongoing observation to see if it spreads nationwide. But overall, I still judge that a complete ban is basically impossible. At most, it will require platforms to disclose more information. Currently, the rebates for market makers in U.S. stocks and options are only disclosed, and there is no hard ban. Moreover, the interests of the two parties are not aligned, so this legislative process will not be quick.
7. A Review of the History of Virtual Asset Regulation in Hong Kong
Hazel: Returning to Hong Kong, I would like to ask Sherry to explain the regulatory logic regarding crypto-related businesses in Hong Kong. You have worked at the Hong Kong Securities and Futures Commission for many years and have participated in the licensing process. Can you review how the licensing approach in Hong Kong has evolved?
Sherry: The Hong Kong Securities and Futures Commission (SFC) actually started regulating virtual assets quite early. During the ICO boom in 2017, many regulatory agencies in other countries were still discussing whether virtual assets were legitimate and whether they warranted legislative regulation. There were even voices suggesting that crypto would eventually disappear and was just a temporary trend.
However, Hong Kong made a judgment at that time: crypto will not disappear, and regulation is unavoidable. The question was—how to regulate? At that time, the SFC did not have the legal authority to issue licenses for such assets. Many other regions used payment licenses to cover "light regulation" from an anti-money laundering perspective.
But Hong Kong is different. The SFC has long-standing experience in regulating traditional finance and has gone through many crises. Therefore, in 2018, it proposed a core concept: "same business, same risk, same regulation." This means that many crypto projects essentially replicate traditional financial behaviors, and since the risks are similar, the regulatory standards cannot be downgraded.
However, the legislative process had not yet started, and the industry was developing rapidly, so the SFC formed two major regulatory strategies in practice, which I refer to as two "comprehensives."
The first "comprehensive" is in-depth regulation. We found that virtual asset trading platforms (what people refer to as crypto exchanges) are the highest-risk segment in the entire ecosystem, especially concerning issues like client asset custody, security, conflicts of interest, and market manipulation. In 2019, without legislative support, the SFC referenced the existing Securities and Futures Ordinance's Type 1 license (securities trading) and Type 7 license (automated trading services) to design a voluntary licensing framework.
This framework is very comprehensive and covers core risk controls in exchange operations, such as client asset custody, cold and hot wallet ratio requirements, cybersecurity, anti-money laundering, and market operation norms. Several institutions became the first batch of platforms to enter the Sandbox.
It wasn't until June 2023 that Hong Kong's Anti-Money Laundering Ordinance officially came into effect, upgrading this framework to a mandatory licensing system.
I remember during those years, we often shared experiences with regulatory agencies around the world, and many countries would ask about the technical standards we established, such as "Why do you require a 98/2 ratio for cold and hot wallets?" and "How is your insurance mechanism designed?" Even today, many regulatory agencies still refer to Hong Kong's institutional framework.
The second "comprehensive" is broad extension. In addition to licensing for exchanges, we also extended to the entire virtual asset business chain, covering brokerages, asset management, advisory services, and various licenses.
For example:
A Type 1 licensed broker can provide virtual asset trading through a comprehensive account and can also act as an introducing agent for exchanges.
A Type 4 licensed advisory firm can provide investment advice on crypto.
A Type 9 asset manager can apply for an upgraded license to manage crypto funds if more than 10% of the funds under management are invested in crypto.
Ivy: You just mentioned regulation related to exchanges. What about businesses involving "stock tokenization"? Are traditional internet brokerages or capital market exchanges allowed to engage in such activities in Hong Kong? What would the regulatory process look like?
Sherry: Hong Kong actually has a very clear restriction in this regard. Section 19 of the Hong Kong Securities and Futures Ordinance grants the Hong Kong Stock Exchange a statutory monopoly. This means that whether on-chain or off-chain, as long as it involves the matching of transactions for Hong Kong stocks, it can only be done by the Hong Kong Stock Exchange.
In other words, any other platform attempting to operate Hong Kong stock trading business (including tokenized forms) is not legally permitted.
So, I was very clear about this during my time at the SFC. Whenever the industry mentions stock tokenization, discussions usually halt at the monopoly position of the Hong Kong Stock Exchange.
This legal status has a long history. If you look back at history, you will know that Hong Kong did not start with just the Hong Kong Stock Exchange. There were once four exchanges, which later consolidated into today's Hong Kong Stock Exchange due to liquidity issues and other factors. This structure is the result of years of evolution, and challenging or questioning it would face significant resistance.
Moreover, this no longer falls within the jurisdiction of the SFC. Changing the monopoly position of the Hong Kong Stock Exchange cannot be pushed solely by the regulatory level; it requires higher-level policy institutions, such as the government and the Legislative Council, to participate in the push.
Nevertheless, within the constraints of the existing regulatory framework, the industry has been actively exploring some "compliant innovative paths." For example, I have seen some institutions attempting to package stocks or other underlying assets into a fund structure and then tokenize them for distribution through a Collective Investment Scheme (CIS). This way, it is no longer directly defined as "stocks," but rather becomes the token issuance of a private equity fund.
From a legal perspective, this approach indeed opens a new path, avoiding the red line of securities matching. However, there are some practical challenges— for example, if a fund wishes to issue to retail investors, it still must obtain approval from the SFC's Investment Products Division, and the approval process is not as efficient as traditional securities issuance.
So currently, this model still has many restrictions in Hong Kong, and progress is somewhat difficult. But I believe that with technological development, the evolution of regulatory concepts, and the gradual realization of "everything on-chain," changes may occur in the future.
Personally, I am very much looking forward to the day when traditional exchanges face the impact of on-chain transformation. At that time, the Hong Kong Stock Exchange may not necessarily have to stick to the existing model; it could very well make a choice.
8. Envisioning the Future: Everything on Chain
Hazel: I think in this last part, we can discuss some more futuristic ideas. Most of what we've talked about so far revolves around the capabilities of traditional stocks, such as the returns from price increases. But will tokenized stocks have some new capabilities in the future? I remember in a podcast episode with Professor Mindao, he mentioned an interesting point: if I hold tokenized stocks of Disney, could I get a discount on tickets when I go to Disneyland in the future? In other words, if stocks are given more "additional rights," will that enhance their attractiveness? Have you thought about what new value tokenized stocks could provide to users beyond traditional shareholder rights like dividends and voting?
Sherry: When I think about this question, I always feel that the real charm of going on-chain lies in its programmability.
In the traditional financial system, securities are "dead." For example, dividends, stock splits, distributions, and voting rights are all handled through announcements, manual processes, and procedures, which are inefficient, inflexible, and provide a poor experience.
But on-chain securities are "alive," and can be dynamically adjusted through smart contracts. For instance, different classes of shares can have different rights, and if a company suddenly wants to provide a benefit to shareholders, such as ticket discounts or exclusive pre-sales, it can be issued in real-time through on-chain contracts without needing to run announcements or manual registration processes. This gives tokenized stocks richer interactive capabilities than traditional stocks and provides users with a more "emotional" and "consumer-oriented" experience.
On the other hand, the future is an AI-driven era, where many IP contents, NFTs, and creator assets will also go on-chain. These assets can interact through smart contracts. The integration of securities with NFTs, identity credentials, and content copyrights will create a very imaginative structure. Although I am not from a technical background and cannot say that these things can be realized immediately, I believe this is a direction worth investing long-term energy in.
Hazel: I understand. So, what do you think, Di? Do you have any special thoughts on the endgame of tokenized stocks or even "everything on-chain"?
Zheng Di: Actually, this question was raised by Michael Saylor back in February this year. When he was conducting research with institutions, he made it very clear: although he didn't use the term "everything on-chain," the meaning was very clear—"the future is the era of the digital economy, where an IP, a house, a collectible, a stock, everything should be able to be traded on-chain."
He said that besides Bitcoin, the overall scale of other digital assets will grow from the current $1 trillion to $590 trillion in the future.
So the on-chain assets we are currently working on, such as money market funds and stocks, actually still belong to the "infrastructure" of traditional finance. These things could originally be traded off-chain.
But what is more important moving forward is whether we can also move "originally non-tradable" assets onto the chain. I think this is the real new asset class.
Imagine if a simple game like "Sheep Game" could be invested in on-chain, or if the box office of "Nezha 2" could be directly invested through tokens, wouldn't that open up a whole new market? In reality, you can't just invest in the box office of a specific movie, but on-chain, you can completely do that—you could buy the future copyright revenue of a song, buy a share of a short drama's streaming revenue, or even invest in a character trained by an AI model.
These types of assets have characteristics of high volatility, explosive potential, and strong cash flow, essentially resembling a "tradable lottery ticket," and they are real-world assets that can generate income, not just air coins. I believe that if Robinhood users could invest in GameStop, they would definitely invest in such content assets because they essentially enjoy "betting on a hit."
Hazel: I particularly resonate with that. In fact, back in 2017 and 2018, we discussed these ideas, and at that time, many narratives around "chain reform" and "coin reform" seemed very distant, even a bit like air. Without a compliance framework, there was no way to connect on-chain and off-chain, and many projects ultimately ended without results.
But now, six or seven years later, as we return to this narrative, it seems that we can really start to realize it this time. One more point I want to add is that the things Zheng Di just mentioned—those that "cannot be traded in reality but everyone wants to trade"—have actually been partially absorbed by prediction markets.
For example, whether Zelensky will wear a suit tomorrow or whether the box office of "Nezha" will break 2 billion, you can place bets on those. But these prediction markets are not tied to the underlying assets, and even if you win the bet, you won't receive box office revenue.
So the truly imaginative scenario is to allow these real-world assets that can create cash flow to also have the ability to be securitized, tradable, and combinable. This could be the ultimate vision of "everything on-chain."
Hazel: We just talked about the tokenization of spot stocks, and it’s easy to think of two directions following this logic:
One is to push forward: the on-chain market for Pre-IPO (pre-IPO) equity;
The other is to push backward: including futures, contracts, and other derivatives markets going on-chain.
Take Pre-IPO as an example; the U.S. has actually had similar practices for a long time. After Robinhood launched tokenized stocks of OpenAI, it also sparked quite a bit of controversy. What are your thoughts on this?
Sherry: From the perspective of Hong Kong regulation, we must return to a key phrase: "same business, same risk, same rules." In the traditional financial system, Pre-IPO equity essentially belongs to private equity, and its sales targets must be limited to professional investors.
Now, if you add a technological wrapper to it, such as packaging it with tokens, from the perspective of the Hong Kong SFC, it still does not change its essence. You cannot bypass regulation just because you used a tech wrapper.
Under this judgment standard:
First, you must certify the client as a professional investor;
Second, you must conduct risk assessments and suitability analyses, such as filling out questionnaires and assessing investment experience, etc.
So in earlier years, when people were doing tokenized Pre-IPOs, they thought that going on-chain would increase liquidity. But the Hong Kong SFC has emphasized efficiency, transparency, and reducing intermediary friction, but has never mentioned "enhancing liquidity." Why? Because truly high-quality assets will be fiercely sought after whether they are on-chain or not; while those assets that no one is interested in will not naturally become more liquid just by going on-chain.
Hazel: That's particularly interesting. "Good assets on-chain are still good assets; bad assets on-chain are still bad assets," the essence does not change. That might also explain why the efficiency of fundraising through tokens in the crypto space has not been high over the years; good projects do not necessarily adopt the token model, while bad projects find it easier to raise money through marketing.
However, I am also thinking, if regulation allows, will traditional startups in the future also mimic the Web3 model, initially using airdrops to attract users and later issuing equity tokens? Essentially retracing the old path of DeFi?
Sherry: I think this path is already very difficult to walk. Because regulation has caught up, the compliance framework has basically been established, and it is not easy to "grow wildly" like in the past.
Hazel: So, Di, what do you think about Robinhood launching OpenAI equity trading? Especially regarding the regulatory boundaries of this "semi-primary market" model for Pre-IPO transfers?
Zheng Di: There is no definitive conclusion on this matter, but it is indeed very worthy of attention. I have also discussed this with Sherry.
From an off-chain perspective, Pre-IPO equity transfers already exist; it's just that their efficiency is very low. When you go on-chain, it becomes high-frequency trading 24/7, which will directly impact the original financing logic of these companies.
The primary market for financing is originally a "competitive game": shout a valuation, circle a group of friends, and once the atmosphere is right, the investment is made. But if you form a continuous trading secondary market on-chain, that valuation will immediately need to be tested. Especially in scenarios where prices are visible and deep, how will auditors and valuators explain your valuation?
This will directly impact the financing logic. Such transactions will shift from "verbal valuation" to "real market testing." For example, a blockchain data company I follow had a valuation of $9 billion two years ago, but the semi-primary market traded at $2.6 billion; how do you expect them to explain that? How will auditors view it? How will fund investors view it?
These companies certainly do not want shareholders to list their shares for trading, but you cannot control that. Whether shareholders can transfer their shares depends on the articles of association and board authority, not the law. Many projects, in order to raise funds, have no ability to restrict the circulation of old shareholders.
To take a step back, even if you write in the regulations that "transfer is not allowed," shareholders can still sign revenue rights agreements and list a contract to create a CFD (Contract for Difference), which cannot be stopped.
Zheng Di: This time, Robinhood was investigated by Lithuania because it used Lithuania's MiFID II investment license, but the Lithuanian regulators only checked "do you have actual stock collateral?" We are not even sure if the OpenAI stocks provided by Robinhood have actual underlying assets.
It could just be a derivative, a non-redeemable contract. If it is indeed just a derivative contract, how will regulation handle it? This is actually a symbolic issue.
I tend to believe that no matter how you circumvent it or how regulation pressures it, this market is "impossible to shut down." Just like the prediction market. More than a decade ago, the U.S. government cracked down on political futures and shut down Intrade, but later PolyMarket emerged. The demand for such markets has always existed; when one platform is shut down, the next batch will emerge.
Every day, people in my social circle are selling old shares of SpaceX, ByteDance, and OpenAI, and there are even shares indirectly linked to ETF models. These transactions have long existed; they are just low in liquidity and efficiency. Going on-chain simply makes it more transparent and efficient. Before strong regulatory crackdowns occur, this trend cannot be stopped.
Even if you do not allow equity trading, there are still prediction markets where you can open a perpetual contract for "OpenAI IPO price prediction" and trade it all the way to the day of the listing. How can you ban that?
9. Guest Q&A Session
Ivy: Our program's traditional segment is here—before we officially conclude, we will ask the guests to pose a question about today's topic, which we may throw to other guests for further discussion in future episodes.
Sherry: I have two questions:
What are everyone's predictions for the next cycle? This cycle's theme is "integration of stocks and coins," so what about the next cycle? Where will the funds flow? Will it continue to integrate or will it re-divide?
What systemic risks might arise from stock tokenization? For example, in terms of technology, contract security, and financial structure, is there a possibility of "踩雷" (踩雷 means to hit a landmine or face unexpected losses) events occurring in the next few years, or triggering new financial risks?
Zheng Di: I also consider this a long-term thinking question. Currently, the push for stock tokenization is primarily driven by second- and third-tier exchanges, while first-tier exchanges are still observing. But whether first-tier or second-tier, everyone shares a common point: they are all very anxious.
The root of this anxiety is quite clear—everyone has been looking forward to new traffic and new users. Now that Web2 users have really come in, they are following the path of internet brokerages rather than exchanges.
If this trend continues, does it mean that internet brokerages will become the biggest winners in this round of stock-coin integration?
Then the question arises: How will CEX (centralized exchanges) respond to this situation? Especially in the next 3 to 5 years, they may really face competition from Nasdaq, the New York Stock Exchange, and the Hong Kong Stock Exchange.
So what I want to ask is: in the face of this potential direct conflict, where is the route for CEX? Can they find new value points instead of relying solely on coin prices and new listings?
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
