Orderly Deleveraging in Progress: A Deep Dive into Crypto Lending and Futures Market in Q2 2026

Lending scale shrinks by $11.3 billion, DeFi lending declines for the third consecutive time, and futures OI rebounds from the bottom.

Written by: Zack Pokorny, Galaxy

Compiled by: Saoirse, Foresight News

As the market continues its deleveraging trend, the second quarter of 2026 marks the first quarter since Q4 2022 where all categories of on-chain lending have declined simultaneously, covering centralized finance (CeFi), decentralized finance (DeFi), and the crypto collateral portion of collateralized debt positions (CDP) stablecoins.

A significant difference between this round of market conditions and the previous bear market is that outstanding loans have decreased steadily and in a stepwise manner, rather than collapsing all at once. In Q2 2022, the crypto collateral lending industry plummeted by over 55%, followed by further declines of 9% and 29% in Q3 and Q4 2022, respectively. In contrast, during this deleveraging cycle, the market has seen declines of only 10%, 5%, and 17% over three consecutive quarters.

We believe that this moderate downward pace represents a healthier deleveraging cycle: the driving factor is the market's proactive reduction of risk, rather than large-scale forced liquidations or counterparty defaults. If lending scales continue to contract in the future, it is expected to maintain this stepwise downward pattern, avoiding a repeat of the massive chain losses seen in 2022.

The deleveraging at the corporate treasury level primarily stems from Strategy's completion of a $1.5 billion debt buyback in May 2026. This move aims to reduce the debt scale of digital asset treasury strategies to $16.1 billion, roughly returning to the debt levels of such companies in July 2025.

In the futures market, the overall change in open interest (OI) at the end of this quarter was minimal, with a slight decrease of 3.08%, totaling $103.2 billion. Beneath this overall slight decline, there was a clear internal structural differentiation: Bitcoin's open interest fell by 6.24% to $45.04 billion, while Ethereum's open interest saw a larger drop of 26.31%, down to $21.99 billion. By the end of the quarter, Bitcoin and Ethereum together accounted for 65% of the total open interest in futures. Notably, this quarter-end stability did not persist; by the end of July, the open interest in futures rebounded to approximately $114 billion, with Bitcoin (around $48 billion) and Ethereum ($25.74 billion) both bouncing back from their Q2 lows.

Key Points

• Overall, the scale of crypto asset collateral lending in Q2 2026 contracted by $11.33 billion (‑16.78%), falling to $56.16 billion, a decrease of 40.13% from the peak of $78.69 billion in Q3 2025.
• The dollar-denominated outstanding loans in DeFi lending applications have declined for the third consecutive quarter, decreasing by $7.79 billion (‑27.61%) this quarter to $20.43 billion.
• According to Galaxy Research, the outstanding debt used by enterprises for direct purchases or to supplement digital asset treasury strategies is $16.1 billion.
• The open interest (OI) in futures, including perpetual contracts, decreased by 3.08% to $103.2 billion.

Crypto Asset Collateral Lending

The following market overview chart illustrates the history and current major participants in the CeFi and DeFi crypto lending sectors. Due to the sharp decline in crypto asset prices and liquidity exhaustion, some of the largest CeFi lending institutions collapsed successively between 2022 and 2023, marked in red in the chart.

Centralized Finance (CeFi)

The table below compares various CeFi lending institutions covered in this market analysis. Some institutions offer diverse services to investors; for example, Coinbase primarily operates as an exchange but also provides credit to users through over-the-counter crypto loans and secured financing. This analysis only accounts for the balance sheet scale of each institution's crypto collateral lending.

As of June 30, the outstanding loans in CeFi, as tracked by Galaxy Research, amounted to $22.98 billion, a decrease of 9.62% (down $2.45 billion) compared to the previous quarter. Compared to the bear market low of $6.8 billion in Q4 2023, it has increased by $16.14 billion, a growth of 235.94%; however, it is still 37.16% lower than the historical peak of $36.58 billion in Q1 2022.

The total scale of CeFi lending contracted in Q2, primarily due to the decline in outstanding loans secured by Tether; Galaxy, Coinbase, Ledn, Arch, Sygnum, and Milo all achieved growth in their lending balance sheets this quarter.

Tether remains the absolute leader in the CeFi lending market, holding a market share of 58.54%, down 371 basis points. Along with Maple (8.91%, up +52bp) and Nexo (7.51%, up +49bp), the top three institutions in the statistics account for a total of 74.96% of the market, with an overall share decrease of 270 basis points.

When comparing market shares, it is important to note that there are significant differences between CeFi institutions: some only offer specific types of loans (e.g., only accepting BTC collateral, only supporting altcoin collateral, or providing cash loans in fiat rather than stablecoins); some only serve specific customer groups (institutional clients / retail users); and their operations are subject to jurisdictional restrictions. These factors lead to varying expansion capabilities among different institutions.

The table below outlines the data sources for each CeFi institution tracked by Galaxy Research and the logic behind the balance sheet scale calculations. Data for DeFi and on-chain CeFi can be obtained from public chain data, which is transparent and easy to read; however, obtaining CeFi data is quite challenging: the accounting standards for outstanding loans vary among institutions, and the frequency of information disclosure is inconsistent, making it difficult to access comprehensive data sources.

Note: Data provided by private third-party institutions has not yet been formally verified by Galaxy Research.

Centralized Finance vs. Decentralized Finance Lending

The dollar-denominated outstanding loans in DeFi lending applications have decreased for the third consecutive quarter, down by $7.79 billion (‑27.61%) to $20.43 billion.

When combining DeFi applications with CeFi lending platforms, the total outstanding crypto collateral loans at the end of the quarter amounted to $43.41 billion, a decrease of $10.24 billion (‑19.08%), primarily driven by on-chain borrowing. Notably, this marks the first time since Q3 2023 that the outstanding loans in CeFi have exceeded those in DeFi lending applications.

Note: There is a risk of double counting between the total scale of CeFi balance sheets and DeFi loan statistics. Some CeFi institutions may lend to off-chain clients using DeFi protocols. For example, a CeFi institution may pledge idle BTC, borrow USDC on-chain, and then lend the USDC to off-chain borrowers. This loan would be counted as both DeFi outstanding loans and as a loan on the institution's financial statements. Due to the lack of disclosure and on-chain identity markers, it is challenging to filter out such double counting.

The contraction in DeFi lending is greater than that in CeFi, leading to the disappearance of DeFi's previous scale advantage. By the end of Q2 2026, the market share of DeFi lending applications fell to 47.05%, a decrease of 555 basis points; the share was 52.6% at the end of Q1.

The third major segment, the crypto collateral portion of collateralized debt positions (CDP) stablecoins, saw a quarter-on-quarter decline of $1.09 billion (‑7.86%). There is also a risk of double counting: some CeFi institutions may obtain funds by minting CDP stablecoins and then lending to off-chain clients.

Overall, crypto asset collateral lending contracted by $11.33 billion (‑16.78%) in Q2, falling to $56.16 billion, a decline of 40.13% from the peak of $78.69 billion in Q3 2025.

Market share breakdown at the end of Q2 2026:

• DeFi lending applications: 36.37% (down ‑544bp)
• CeFi lending platforms: 40.93% (up +324bp)
• Crypto collateral portion of CDP stablecoins: 22.7% (up +220bp)

When combining DeFi lending with the crypto collateral portion of CDP stablecoins as on-chain lending, the total market share is 59.07%, a decrease of 324 basis points.

More Perspectives on Decentralized Finance Lending

The outstanding loans of DeFi lending applications have continued to shrink since reaching a historical high of $47.13 billion on September 19, 2025; as of July 21, 2026, the scale was $21.94 billion, a decrease of $25.19 billion, or 53.45%.

Since the end of the first quarter of 2026, the pullback in DeFi borrowing has intensified, but there have been recent signs of slight easing.

Stablecoins

According to the 7-day moving average, from March 31 to June 30, the weighted average borrowing rate for stablecoins rose by 27 basis points this quarter; after the quarter ended, the rate continued to rise to 3.88%.

This indicator comprehensively calculates the borrowing costs of lending protocols and the CDP stablecoin minting fees, weighted by the outstanding loan scale.

The chart below breaks down two types of costs: borrowing stablecoins through lending protocols and minting CDP stablecoins with crypto assets as collateral. The interest rate trends of both are highly correlated, but the CDP minting rate fluctuates less, as this rate is set manually and does not change with market fluctuations. For more than 21 months, both rates have been supported at the bottom by the U.S. federal funds rate.

In the second quarter, the OTC benchmark rate for USDC was in the range of 4.25%-5%; it remained at 4.25% at the end of the quarter and continued at this level until August 3.

The OTC lending rate for USDT also fluctuated in the range of 4.25%-5%.

Bitcoin

The chart shows the weighted borrowing rates of wrapped Bitcoin (WBTC) across multiple public chain lending applications. On-chain WBTC is mostly used as collateral, and the borrowing demand is not strong, resulting in long-term low borrowing costs. Unlike stablecoins, the on-chain borrowing rate for BTC is very stable, with low frequency of borrowing and repayment by users. In the second quarter, the on-chain BTC borrowing rate fluctuated between 0.44%-0.5%.

The historical price difference between on-chain and off-chain (OTC) borrowing rates for BTC continued to exist this quarter. The demand for BTC borrowing in the OTC market comes from two points: ① the demand for shorting Bitcoin; ② using BTC as collateral to borrow stablecoins or fiat currency. The demand for shorting is not common in the on-chain lending market, which creates a cost price difference between on-chain and off-chain.

The off-chain BTC rate remained unchanged at 1% this quarter.

ETH and stETH

The chart below shows the weighted borrowing rates for ETH and stETH (staked Ethereum generated by the Lido protocol) across various lending protocols and public chains. Historically, the borrowing cost for ETH is higher than for stETH due to stronger borrowing demand for ETH.

Users frequently borrow ETH to execute a circular leverage strategy: staking stETH (the certificate obtained from staking ETH with Lido), borrowing ETH, and obtaining leveraged exposure to the annualized yield from Ethereum staking. Under normal market conditions, the borrowing cost for ETH fluctuates within 50 basis points around the Ethereum staking APY. Once the borrowing cost exceeds the staking yield, this strategy becomes unprofitable, making it difficult for borrowing APY to remain consistently above the staking yield.

Similar to WBTC, stETH is mostly used as collateral, so the cost of borrowing stETH is usually very low.

Users can use liquid staking tokens (LST) and liquid re-staking tokens (LRT) as collateral to borrow ETH at very low or even negative net rates. This gives rise to a classic circular strategy: repeatedly depositing LST/LRT as collateral, borrowing unstaked ETH, then staking ETH to obtain new LST/LRT, and continuing to borrow more ETH to amplify staking yield exposure. The premise for this strategy to work is that the borrowing cost for ETH is lower than the annualized yield obtainable from LST/LRT staking. Except for a few special periods, this strategy can generally operate smoothly.

Analysis of Aave Lending Ledger

The following is a deep analysis of the Aave V3 core instance (the largest on-chain lending market) ledger after applying three filtering rules:

• Minimum debt threshold ($100): Summary statistics exclude positions below this threshold, filtering out small positions, but this skews the statistics towards larger loans.
• Health factor HF report upper limit (HF ≤ 50): Positions with a health factor greater than 50 are not included in the core statistics; these over-collateralized positions are usually small in amount and not significant for risk analysis. Within the threshold, the debt-weighted average health factor and quantile statistics only include positions where 1 ≤ HF ≤ 50; positions with HF < 1 are not included in health factor statistics, while those with HF = 1 are included in the statistics.

The higher the health factor, the safer the position; a health factor below 1 indicates that the position triggers liquidation conditions. The health factor calculation formula = (total value of collateral × weighted average liquidation threshold) ÷ total loan amount.

• Debt-to-equity ratio D/E: Only positions where the value of collateral > debt (net assets are positive) are counted; positions with negative net assets are excluded from the D/E distribution and debt-weighted average D/E. This rule is independent of the $100 debt threshold: even if a loan exceeds $100, if there are no positive net assets, it is also excluded from the D/E statistics.

Based on the snapshot of August 7, 2026, there are 19,073 valid outstanding loans after filtering. The number of loans in the "efficiency mode (e-mode)" accounts for only 8.91% of total positions, but the outstanding debt scale is almost 50% of ordinary loans. The characteristic of e-mode is that the borrowing asset is highly correlated with the collateral asset price (e.g., borrowing WETH against ETH collateral). The last Galaxy statistic on April 22 showed that the debt ratio of e-mode was close to 60:40; the decline in proportion is due to the decrease in outstanding debt scale in e-mode.

The table below shows the debt-weighted risk indicators of the filtered ledger, divided into three groups: all positions, e-mode, and ordinary mode. e-mode borrowers have extremely high overall leverage: the debt-weighted loan-to-value ratio (LTV) is about 90%, the debt-weighted health factor is only 1.06, and the debt-to-equity ratio (D/E) is about 10.7; this means that even a slight shock to the collateral price will push a large number of positions into the pressure zone.

Ordinary mode loans have a much thicker safety cushion: the debt-weighted LTV is about 49%, the health factor is 1.79, and D/E is about 1.07; this is used to hedge risks unrelated to the price of collateral and borrowing assets, such as borrowing USDC against cbBTC collateral.

The debt-weighted D/E calculation formula (only for Ci > Di, i.e., collateral > debt positions): D/E = Σi (Di × (Di ÷ (Ci − Di))) ÷ Σi Di, where Di represents the debt of a single position, and Ci represents the value of collateral for a single position.

Next, we will analyze the dollar value proportions of various collateral assets in the Aave V3 core market. ETH-based collateral dominates: WETH accounts for about 24%, weETH (wrapped tokens for re-staked ETH by Etherfi) accounts for 16%, and wstETH (wrapped tokens for Lido stETH) accounts for 14%, together making up 54.6% of all available collateral; WBTC accounts for about 14%. A few assets bear the majority of the collateral value on the books, with the remainder being stablecoins and other yield-bearing tokens.

Looking at the composition of borrowing assets: WETH accounts for slightly over 37% of total liabilities, which is a typical result of the prevalence of ETH-based collateralized leverage strategies. Stablecoin borrowing is also substantial: USDT accounts for about 28%, and USDC accounts for about 22%, together making up half of the total borrowing scale, while other cryptocurrencies account for a smaller proportion.

Compared to the last statistics, the proportion of WETH in outstanding liabilities has significantly decreased from 51.1%, consistent with the previously mentioned e-mode debt contraction.

e-mode Detailed Perspective

e-mode's collateral assets are highly concentrated in ETH staking/re-staking wrapped tokens: weETH alone accounts for 42% of this category's collateral; combined with rsETH and wstETH, these three assets account for 66.2% of e-mode collateral. This indicates that the risks in e-mode are not diversified collateral but essentially concentrated bets on the fundamentals of Ethereum staking.

On the borrowing side, the vast majority of e-mode debt is denominated in WETH, which alone accounts for 73% of e-mode debt, fully aligning with users' behavior of borrowing ETH using ETH-based collateral. Stablecoins still hold a certain proportion, with USDT, USDe, and USDC collectively accounting for about ten percent of e-mode borrowing scale.

The table below summarizes the e-mode positions corresponding to each collateral asset, including debt-weighted risk indicators; for sub-samples where 99% of collateral is concentrated in a single asset, implied leverage multiples will also be calculated. This serves as a ranking of "which asset is leveraged the most aggressively," differing from ordinary market capitalization ranking tables. Liquid staking and re-staking ETH tokens rank at the top, with high debt-weighted LTVs, D/E ratios generally in the high single digits to teens, and health factors slightly above 1, corresponding to a large number of tightly-coupled ETH leveraged positions.

The implied leverage multiple for a single position is calculated using the formula: N_i = ln((1 − (D/E)_i(1 − Li)) / Li) / ln(Li), where Li is the LTV of that position; this is only applicable for sub-samples where collateral is ≥99% in a single asset. The output is the debt-weighted average Ni; positions where L is not in (0,1), the logarithmic parameter is non-positive, or the result is not a finite value are excluded. Values are output only for liquid staking ETH, liquid re-staking ETH, yield-bearing stablecoins, and Pendle PT tokens.

Corporate Debt Strategies

Galaxy Research is currently tracking $16.1 billion in outstanding debt, which corporations are using to directly purchase or supplement digital asset treasury strategies. Due to Bloomberg's limitations in tracking the preferred stock statistics of Strategy Company, the time series of incremental STRC circulating shares is misaligned, but the total debt scale accurately reflects actual liabilities.

Strategy completed a $1.5 billion debt buyback in May, and the outstanding debt of digital asset treasury companies (DAT) decreased by $1.5 billion this quarter.

The table below shows the actual interest payments required each quarter for DAT-issued debt. Note that Strategy's STRC dividends must be approved by the board and distributed from legally available funds; unpaid dividends will continue to accumulate and must be prioritized for payment before any profits can be allocated to junior securities. Therefore, the timing of STRC dividends is irregular, and the amounts are unevenly distributed.

Including DAT corporate debt, the total outstanding crypto-related debt across the industry decreased by 15.08% this quarter. Following a historical high in Q3 2025, on-chain and off-chain crypto-related debt fell to $73.2 billion by the end of Q2, marking three consecutive quarters of decline.

Futures Market

Including perpetual contracts, the open interest (OI) in futures decreased by 3.08% to $103.2 billion; the open interest in July rebounded, reaching about $114 billion by the end of the month.

Important Note: The open interest does not equate to the absolute total leverage. Some open positions may be hedged by spot longs, achieving delta-neutral exposure; OI alone cannot directly indicate the overall market leverage ratio.

In Q2, BTC futures open interest fluctuated between $44 billion and $62 billion; it started at $48.04 billion and dropped to $45.04 billion by June 30 (-6.24%). After the end of Q2, it rebounded to around $48 billion in early August.

ETH's futures open interest saw a larger decline than BTC in Q2: starting at $29.84 billion, it fell to $21.99 billion by June 30, a decrease of 26.31%; after the end of the quarter, it rebounded to $25.74 billion.

By the end of Q2, the combined open interest of BTC and ETH was $67.07 billion, accounting for 65% of the entire futures market.

Conclusion

We believe that following the significant drop in the futures market on October 10, 2025, the second quarter of 2026 further confirms that leverage in the crypto market is gradually being digested. The lending market is experiencing a stair-step decline rather than an elevator-style crash, with three consecutive quarters of moderate contraction instead of a repeat of the cliff-like drop seen in the 2022 bear market.

Corporate treasury debt and futures open interest also reflect a controlled decline rather than a forced deleveraging. Data from early July has already shown that futures open interest and DeFi borrowing scales may be nearing a bottom range.

If this trend continues, the market will have a stronger capacity to withstand further contraction, likely avoiding the chain liquidations and cascading defaults seen in the previous cycle. As it stands, deleveraging is still progressing step by step.