I. Executive Summary

Pendle’s Boros separates funding rates into fixed and floating legs. In essence, it recreates the same structure used in traditional interest rate swaps: pay fixed and receive floating—or vice versa.

Our key conclusions on cross-exchange funding rate swaps using Boros are as follows:

1. Boros eliminates the primary loss scenario in traditional funding-rate arbitrage: funding-rate reversal

By locking in the fixed leg through Boros and hedging the floating leg with perpetual futures, traders can remove their exposure to changes in realized funding rates.

Boros charges:

• A 0.05% position fee each time a position is opened
• A 0.2% annualized settlement fee, calculated on notional exposure and charged every eight hours

Actual margin posted on Boros is less than 1% of notional exposure. YU positions referencing the same underlying asset can also share collateral within a Cross-Collateral Zone, allowing their mark-to-market movements to offset one another.

After fees, all four backtested market-months generated positive net returns:

• Event-driven months: 13–15% annualized net returns
• Steady-state months: 6–9% annualized net returns

Returns improve further after accounting for interest earned on Binance margin collateral.

2. Capacity is limited

The average amount of deployable margin per arbitrage opportunity is only approximately 1.21 BTC.

Each market has an initial open-interest cap of $10 million in notional exposure, with maximum leverage of 1.2x.

This makes the strategy best suited to relatively small amounts of capital seeking high-certainty carry, rather than large-scale deployment.

II. Where Does the Opportunity Come From?

2.1 How Boros Works: Yield Units

Boros operates on Arbitrum and is built around a product called the Yield Unit, or YU.

Each YU represents the funding-rate income stream generated by one unit of an underlying asset on a particular exchange.

For example:

1 YU-BTCUSDT (Binance) represents the funding-rate cash flows earned by 1 BTC of exposure in Binance’s BTC perpetual market.

Each YU has two relevant rates:

• Implied APR: The rate priced by the Boros order book
• Underlying APR: The actual funding rate realized on the referenced exchange

Going long YU means paying the fixed rate and receiving the floating rate.

Going short YU means receiving the fixed rate and paying the floating rate.

At maturity, the difference between the two is settled in cash.

2.2 Four Ways to Use Boros

Use Case 1: A Pure Cross-Exchange Pair Trade Inside Boros

A trader shorts the YU linked to the higher-funding exchange and goes long the YU linked to the lower-funding exchange.

The difference between the two implied APRs becomes the fixed annualized spread locked in by the trade.

This structure requires only Boros margin and therefore offers the highest capital efficiency.

However, it comes with an important risk.

While the spread between the two YUs is locked in, the position is not fully hedged against a simultaneous directional move in both implied APRs.

For example, during a market panic, funding rates across all exchanges may fall sharply at the same time. Both YU positions could then experience mark-to-market losses in the same direction, potentially triggering a margin call.

How significant is this risk?

Underlying APR offers a useful reference point. BTC’s monthly annualized funding rate moved from approximately +55% in January 2026 to −0.68% in February 2026—a swing of more than 50 percentage points in a single month.

Even if implied APR moves only half as much as underlying APR, a position using 1.2x leverage could experience mark-to-market changes equivalent to approximately 30–50% of posted margin.

That is enough to bring the position close to a margin call or liquidation.

Use Case 1 is therefore best suited to short-dated contracts—typically one to two weeks to maturity—and relatively small trial positions.

Use Case 2: A Fully Hedged Four-Leg Structure

This is the structure used in our return calculations.

It consists of four legs:

A. Short Hyperliquid YU on Boros
Receive fixed funding and pay floating funding

B. Short BTC perpetuals on Hyperliquid
Receive floating funding

C. Long Binance YU on Boros
Pay fixed funding and receive floating funding

D. Long BTC perpetuals on Binance
Pay floating funding

Legs A and B together convert the Hyperliquid side into a position that receives a fixed funding rate.

Legs C and D together convert the Binance side into a position that pays a fixed funding rate.

Legs B and D neutralize BTC price exposure.

The combined result is:

Receive the fixed funding-rate spread between the two exchanges, with no directional BTC exposure and no exposure to fluctuations in realized funding rates.

There is, however, an important directional condition.

The structure only works when Hyperliquid’s implied APR is higher than Binance’s implied APR. Otherwise, the portfolio pays the spread instead of receiving it.

This is not a trade that can be entered at any time. It becomes attractive only when the implied APR on the high-funding venue is sufficiently above that of the low-funding venue.

The cost is greater operational complexity.

The four legs are distributed across Boros and two perpetual venues, and each position can be liquidated independently. Pendle’s original backtest assumed 10x leverage on the exchange-side positions.

In practice, the four legs cannot be entered or exited at exactly the same time.

From identifying an opportunity to completing all four trades, execution would realistically take approximately 5–15 minutes:

• Faster when liquidity is strong
• Slower during event-driven volatility

Most execution risk is concentrated in the Boros YU legs, where liquidity is thinner. Slippage on the perpetual legs should be relatively limited because exchange order-book depth is substantially greater than on Boros.

Use Case 3: Converting Floating Carry into Fixed Carry

A trader can short a YU linked to a particular exchange while maintaining a perpetual position on that same exchange.

This converts a floating funding-rate return into a fixed carry return.

The structure is a natural hedging tool for delta-neutral protocols such as Ethena.

Use Case 4: Directional Funding-Rate Trading

Traders can also go long or short YU based on their view of where funding rates are heading.

This is effectively a directional trade on funding rates rather than an arbitrage strategy.

2.3 Why Cross-Exchange Funding Spreads Can Persist

Funding rates for the same asset can remain different across exchanges for extended periods. This is a structural feature of the market rather than a temporary anomaly.

One important quantitative difference is the concentration of open positions.

Hyperliquid currently has approximately:

• $10.2 billion in 24-hour open interest
• $10.5 billion in 24-hour trading volume

Its open-interest-to-volume ratio is therefore close to 1.0x.

In other words, one full day of trading volume is only roughly equal to the value of all outstanding positions. This suggests slower position turnover and a greater concentration of leveraged speculative exposure.

By comparison, Binance’s open-interest-to-daily-volume ratio is generally around 0.2–0.3x.

Open interest is much smaller relative to daily trading volume, positions turn over more quickly, and arbitrage capital can close pricing gaps more efficiently.

Hyperliquid’s system-level leverage ratio of approximately 4x, measured as open interest divided by TVL, provides further evidence of concentrated leverage.

These structural differences make Hyperliquid funding rates more likely to surge when long positioning becomes crowded—and to fall more sharply during sell-offs.

This is the structural origin of the cross-exchange funding spread.

The two exchanges also use different funding-rate formulas, creating natural differences in both absolute funding levels and the timing of funding-rate fluctuations.

Pendle’s backtest supports this view.

Over a two-month observation window, the BTC funding-rate spread between Hyperliquid and Binance produced a weighted-average implied APR spread of approximately 9.94%, with peaks above 23%.

BNB offers even stronger empirical evidence of structural cross-exchange spreads

Binance applies a 0% baseline funding rate to BNB perpetuals.

Observed data over 30 months shows:

• Average Binance BNB funding APR: −1.86%
• Average Bybit BNB funding APR: +1.87%
• Average absolute spread: 8.53% APR

This is substantially higher than the corresponding cross-exchange spreads for BTC and ETH.

The spread remained above:

• The monthly break-even level of 2.4% for 84.3% of the observed period
• The quarterly break-even level of 1.1% for more than 90% of the observed period

BNB therefore appears to offer a more stable underlying market with a wider execution window than BTC or ETH.

III. Cost and Return Analysis

Verified Fees and Protocol Parameters

Fee Breakdown

The Boros settlement fee is an annualized rate of 0.2%, calculated on notional exposure and charged every eight hours.

The formula is:

|Position Size| × 0.2% × (8 hours / 8,760 hours)

Capital Allocation

Assume a four-leg structure with $1 million in notional exposure per leg.

The portfolio has:

• $2 million in total gross notional exposure
• Zero net directional notional after offsetting the long and short positions

Exchange-side margin

• Leg B: $200,000
• Leg D: $330,000
• Total exchange-side margin: $530,000

Boros-side margin

Boros uses the following initial-margin formula:

IM = Notional × Time to Maturity × Implied APR / Leverage

For a one-month contract using 1x leverage, initial margin is only approximately $8,200 per leg, or roughly $16,000 across both legs.

In addition, Binance and Hyperliquid YU positions referencing the same base asset can share margin within a Cross-Collateral Zone.

This allows unrealized gains and losses on Legs A and C to offset one another, significantly reducing net mark-to-market volatility.

The buffer can be estimated as follows:

• Monthly BTC implied-APR volatility: approximately 1.5–2%
• Derived from annualized volatility of roughly 5–7%, divided by √12
• At $1 million in notional exposure, monthly mark-to-market volatility per leg is approximately $15,000–$20,000
• With cross-collateralization, net volatility across the two legs is likely only 30–50% of single-leg volatility
• This implies net monthly volatility of approximately $5,000–$10,000
• A three-standard-deviation move would therefore be approximately $15,000–$30,000

We consequently allocate an approximately $50,000 Boros-side buffer, consisting of:

• $16,000 in initial margin
• Approximately $30,000 for a three-standard-deviation mark-to-market move
• An additional safety margin

Total capital required is therefore approximately:

$580,000

Effective portfolio leverage is approximately:

1.7x

Monthly Fees

Annualized fees are:

$2,033 × 12 = $24,400

Relative to total capital of $580,000, annualized costs are approximately:

4.2%

Longer Maturities Are More Fee-Efficient—but Liquidity Is Currently Close to Zero

Net Returns on Total Capital Across Four Backtested Markets

Distinguishing Steady-State Returns from Event-Driven Returns

For the 2023–2026 BTC subsample:

• Mean funding APR: 7.32%
• Median funding APR: 6.79%

Using monthly contracts, a gross APR of 7.32% translates into an estimated annualized net return of approximately 8.4%.

The return profile can therefore be divided into three broad regimes:

Steady-state environment

• Gross APR: 6–8%
• Net return: 5–9%

Event-driven environment

• Gross APR: 10–15%
• Net return: 10–18%

Extreme negative-funding environment

The strategy becomes inactive.

Negative funding occurred during approximately 14.4% of the BTC subsample, equivalent to roughly two to three months per year.

IV. The Opportunity Window Remains Meaningful

New Underlying Assets Create New Opportunities

Boros has already expanded beyond BTC and ETH to markets including:

• NVDA
• Gold
• Silver

Every newly introduced asset creates a new cross-exchange spread that has not yet been fully arbitraged away.

Funding rates for equity perpetuals are significantly more volatile than those for BTC and ETH, which should create larger initial pricing dislocations.

Macro Volatility Periodically Resets Funding Spreads

Federal Reserve decisions, large-scale liquidations and black-swan events can all amplify differences in funding rates across exchanges.

These event-driven opportunities cannot be permanently arbitraged away.

As long as leveraged speculation exists in perpetual markets, cross-exchange funding spreads will periodically reappear.

Capacity Constraints Can Also Protect Returns

The assumed 0.2-percentage-point price impact per trade is a conservative backtesting assumption, not a physical limit.

Pendle has explicitly noted that traders can consume more order-book depth during periods of unusually wide spreads and accept a less favorable fixed rate.

For example, when the initial spread is 20% APR, traders may still earn attractive returns after filling orders at a fixed APR of 15–16%.

This means that the amount of capital deployable during event-driven opportunities may be greater than the previously estimated approximately 1.21 BTC per trade.

The trade-off is that the realized fixed APR would be lower than the weighted-average rate used in the backtest.

In other words, traders can exchange some APR for greater position size.

For smaller pools of capital, the open-interest cap and limited order-book depth continue to protect returns from excessive competition. However, the strategy’s capacity ceiling is not completely rigid.

The Best Entry Point Is a Transitional Period

The optimal entry window occurs when:

The funding spread has already widened, but market volatility has not yet become disorderly.

This should be identified using market conditions rather than a fixed time interval.

A practical set of entry criteria would be:

• The implied-APR spread has widened to more than twice the break-even level—for example, above 5% for monthly contracts
• BTC’s hourly volatility has fallen by more than 50% from its event-driven peak

Different events—including FOMC decisions, CPI releases and on-chain liquidations—produce very different patterns of spread expansion and convergence.

Execution should therefore be guided by market conditions, not by a fixed number of hours or days after an event.

About BlockBooster

BlockBooster is a next-era alternative asset management firm for the digital age. The firm leverages blockchain technology to invest in, incubate, and manage the core assets of this new era, from Web3-native projects to real-world assets (RWA). As value co-creators, BlockBooster is dedicated to unlocking the long-term potential of these assets, capturing exceptional value for its partners and investors in the digital economy.

Through BlockBooster Research, we share selected insights from our ongoing work across investment, incubation, and alternative asset management, helping readers track the market structures and asset categories shaping the next generation of finance.

Disclaimer

This article/blog is provided for informational purposes only. It represents the views of the author(s) and it does not represent the views of BlockBooster or its affiliates. It is not intended to provide (i) investment advice or an investment recommendation; (ii) an offer or solicitation to buy, sell, or hold digital assets, or (iii) financial, accounting, legal, or tax advice. Digital asset holdings, including stablecoins and NFTs, involve a high degree of risk, can fluctuate greatly, and can even become worthless. You should carefully consider whether trading or holding digital assets is suitable for you in light of your financial condition.

Please consult your legal/tax/investment professional for questions about your specific circumstances. Information (including market data and statistical information, if any) appearing in this post is for general information purposes only. While all reasonable care has been taken in preparing this data and graphs, no responsibility or liability is accepted for any errors of fact or omission expressed herein.