The race in tokenization is shifting. Giving traders access to stocks, commodities and other real-world assets on a crypto exchange is no longer enough: the next step is making those assets work as capital. A new report by Block Scholes and Bitget, published on October 7, 2026, examines how Bitget’s Cross-Asset Unified Account (UTA) does this and what it means for capital efficiency and risk.
From separate accounts to one collateral pool
Bitget serves more than 125 million users and positions itself as a Universal Exchange (UEX), combining the strengths of centralized and decentralized platforms. Through one account, users can trade over 2 million crypto assets alongside tokenized stocks, ETFs, commodities, precious metals and forex.
In July 2026, Bitget launched the Cross-Asset Unified Account, the next stage of its UEX product. It brings tokenized US stocks into the same margin framework used for crypto trading. More than 370 eligible assets, including 125 tokenized US stocks (rStocks), now sit in a single margin pool. Supported names include Apple (rAAPL), Amazon (rAMZN), Google (rGOOGL), Nvidia (rNVDA) and the Nasdaq-100 ETF (rQQQ).
The account evolved in three generations:
| Generation | How margin works |
|---|---|
| Generation 1 | Spot, margin and futures sit in separate wallets; each position and asset is margined on its own |
| Generation 2 | All crypto assets merge into one shared margin pool that backs every position |
| Generation 3 (Cross-Asset UTA) | Tokenized stocks and other real-world assets join the same pool as collateral |
A $1M portfolio: $165K less margin
To measure the gain, Block Scholes modeled a hypothetical $1M institutional book with a long crypto bias and an AI-chip versus Nasdaq-100 relative-value sleeve:
| Position | Size | Leverage |
|---|---|---|
| Long Nvidia, AMD, Broadcom, TSMC and Micron rStocks | $175K | Spot |
| Long BTC perpetual | $410K | 5x |
| Long ETH perpetual | $300K | 5x |
| Short Nasdaq-100 ETF perpetual | $115K | 5x |
In August 2026 the portfolio returned 17% on its $1M notional, driven mostly by the rally in BTC and ETH.
Under generation 1, the trader would pay $175K for the stocks plus $165K of USDT margin for the perpetuals ($82K for BTC, $60K for ETH and $23K for the Nasdaq-100 short). Generation 2 required the same $165K, but in one shared pool, so gains on one leg could offset losses on another.
Under generation 3, the $175K of chip stocks count toward collateral at a 95% discount rate, contributing about $166K. That fully covers the $165K margin requirement, so the trader posts no extra USDT. The only capital committed is the $175K spent on stocks the trader wanted to hold anyway. That is $165K less, or 16.5% of notional and roughly half the capital otherwise required.
How the discount rate works
The discount rate is the share of an asset’s market value that the account counts as collateral. Stablecoins are the benchmark at 100%. Every other asset is discounted to allow for its value falling while a position is being liquidated.
| Asset tier | Discount rate (small positions) |
|---|---|
| USDT, USDC | 100% |
| BTC, ETH, BGUSD | 98% |
| Other stablecoins (USDe, PYUSD) | 95% |
| Large-cap altcoins and large US stocks | 95% |
| Mid-cap altcoins (UNI, AAVE) | 90% |
| Smaller tokens (OP, POL) | 80% |
Two factors set the rate. The first is the asset itself: more liquid, less volatile assets contribute more. BTC and ETH are discounted less than large-cap stocks because they trade around the clock in deep markets, while tokenized stocks can be thinner outside traditional market hours. Bitget still trades rStocks continuously, including weekends and holidays, so they can be liquidated at any time against the mark price.
The second factor is position size. Larger holdings get a lower rate, reflecting the market impact of unwinding them. For BTC and ETH, the rate stays at 98% below $1M and steps down gradually to 50% between $80M and $100M. A large-cap tokenized stock such as rNVDA holds 95% up to about $500K before stepping down, while more volatile names such as rMSTR decline faster.
Stress test: the collateral buffer
The modeled portfolio opens with adjusted equity of about $166K, well above the maintenance margin of about $8,150 at which liquidation begins. Because the account starts close to its initial margin, Block Scholes tested how the buffer behaves under stress.
| Scenario | −10% | −20% |
|---|---|---|
| Chip stocks fall alone | $150K | $133K |
| Chips, BTC, ETH and Nasdaq-100 fall together | $90K | $14K |
A sell-off limited to chipmakers has a contained effect, since the stocks are held without leverage. A broad sell-off hits harder, mostly through the leveraged BTC and ETH legs: of the $76K drop at −10%, $16.6K comes from the stock collateral.
On this book, liquidation is reached at about a 21% correlated decline. Roughly $30K of additional USDT would extend that to about 25%. Holding $166K of USDT instead of stock collateral would push the liquidation point to about 27%. The report’s conclusion: the choice of collateral matters as much as its amount.
How correlation affects capital efficiency
Crypto and equities share common drivers such as global liquidity, real rates and risk appetite, so positive correlation between them is the most common state. Since January 2022, the 60-day correlation between BTC and the Nasdaq-100 ETF has averaged +0.41, ranging from −0.13 to +0.75. It dropped to an average of +0.07 between November 2023 and May 2024, then rose again and averaged +0.53 during the April–June 2025 tariff shock.
Volatility is the second factor. Since 2022, BTC has averaged 51% annualized volatility versus 22% for the Nasdaq-100. Lower-volatility collateral helps preserve the margin buffer. However, the chip basket ran at 56% volatility from June 1 to August 25, 2026, above BTC’s 40%, while the Nasdaq-100 ETF stayed at 26%. Diversified indices therefore show the lower-volatility benefit more consistently than single stocks or a concentrated sector basket.
Block Scholes maps collateral on a grid of correlation and volatility. Low-volatility, negatively correlated collateral carries the lowest relative liquidation risk, and high-volatility, positively correlated collateral the highest. As of August 25, 2026, the chip basket sat just inside the highest-risk cell, with +0.19 correlation to the book and 56% volatility. USDT sits in the low-risk cell.
One rStock, several roles
In the unified account, a single rStock holding can serve several purposes at once:
- Equity exposure: the token tracks the underlying US share.
- Dividends in USDT: Reality, the issuer behind Bitget’s Stocks 2.0, pays eligible cash dividends directly in USDT instead of reinvesting them.
- Margin: the holding can back futures and other positions without being sold.
- Stablecoin borrowing: the holding can be pledged to borrow stablecoins.
Exposure and dividends come from owning the token. Margin and borrowing draw on the same collateral value, so collateral used for one is not available for the other.
Conclusion
Bitget’s Cross-Asset Unified Account extends portfolio margining from crypto to tokenized equities and other real-world assets. In the Block Scholes model, positions that previously required about $165K of separate USDT margin can be backed by stock holdings the trader already owns. The resilience of the book depends on the collateral mix: its correlation and volatility determine how quickly the margin buffer erodes. Bitget can also adjust discount rates dynamically, changing how much margin a holding provides.
