The cryptocurrency market experienced an unusually orderly deleveraging process during the second quarter of 2026, with lending volumes adjusting through a gradual step-down pattern rather than a crash-style plunge. This adjustment spanned centralized finance (CeFi), decentralized finance (DeFi), and the crypto-collateralized segment of collateralized debt position (CDP) stablecoins, marking the first time since Q4 2022 that all major on-chain lending categories declined simultaneously. This non-crash cooling signals that the market is proactively reducing risk to digest previously accumulated leverage, rather than relying on forced liquidations or counterparty failures to passively clear excess.
Overall, crypto-collateralized lending contracted by $11.33 billion during the quarter, a decline of 16.78%, bringing the total to $56.16 billion. Compared to the peak of $78.69 billion reached in Q3 2025, current levels have dropped 40.13%. This data not only reflects a repricing of market liquidity but also reveals a profound shift in risk appetite among both institutional and retail participants. Notably, this moderate downward pace is viewed as a healthier deleveraging cycle, driven primarily by market participants gradually reducing risk exposure, thereby avoiding the cascade of massive losses seen in 2022. If lending volumes continue to contract in the future, this step-down pattern is expected to persist, laying the groundwork for long-term market stability.
The key distinction between this deleveraging cycle and the previous bear market lies in the smooth, staircase-like trajectory of outstanding loan declines, as opposed to a one-time cliff-like crash. In Q2 2022, the crypto-collateralized lending industry plunged over 55% in a single quarter, followed by additional drops of 9% and 29% in Q3 and Q4 2022 respectively, triggering massive forced liquidations of leveraged positions and severely damaging market confidence. In contrast, the current deleveraging cycle has seen declines of just 10%, 5%, and 17% over three consecutive quarters, representing the market's ability to actively and gradually reduce risk. The underlying reasons include a more mature market structure, more cautious leverage usage by participants, and an improved regulatory environment prompting institutions to adopt more conservative risk management strategies. Should lending volumes continue to shrink, this step-down pattern is expected to persist without repeating the chain-reaction losses of 2022.
This difference is evident not only in the data but also in the behavioral patterns of market participants. Proactive risk reduction has become the norm rather than forced responses to liquidations, indicating a transition from extensive growth to refined operations. More critically, this orderly deleveraging process has not triggered liquidity drought; instead, it has provided the market with valuable adjustment time, allowing price discovery mechanisms to function normally.
On the corporate treasury front, the primary driver of deleveraging came from Strategy (MSTR.US)'s $1.5 billion debt repurchase completed in May 2026. This move reduced the debt supporting its digital asset treasury strategy to $16.1 billion, roughly returning to the company's July 2025 debt levels. As one of the largest corporate Bitcoin holders in the crypto market, Strategy (MSTR.US)'s debt management actions carry significant signaling weight for the entire market. The $1.5 billion repurchase not only lightens the company's interest burden but also sends a message of prudent operations to the market. Currently, Galaxy Research tracks outstanding debt used by enterprises to directly purchase or supplement digital asset treasury strategies at $16.1 billion, a decrease of $1.5 billion from the previous quarter. Notably, despite timing mismatches in the issuance of Strategy (MSTR.US)'s STRC circulating shares, the total debt figure accurately reflects actual liabilities. STRC dividends require board approval and must be paid from legally available funds; unpaid dividends accumulate and must be prioritized before distributions to junior securities, resulting in irregular dividend timing and uneven amounts. This strengthened financial discipline enhances the company's resilience to market volatility.
Structurally, the contraction of corporate treasury debt is not an isolated phenomenon but a vital component of the broader crypto ecosystem deleveraging process. As more companies adopt similar debt management strategies, overall market leverage is expected to decline further, reducing the probability of systemic risk events.
In the futures market, open interest (OI) saw minimal change quarter-over-quarter, declining just 3.08% to $103.2 billion by quarter-end. However, beneath this modest overall decline, internal structural divergence was pronounced. Bitcoin open interest fell 6.24% to $45.04 billion, while Ethereum open interest dropped a steeper 26.31% to $21.99 billion. At quarter-end, Bitcoin and Ethereum together accounted for 65% of total futures open interest. This divergence reflects different sentiment dynamics across asset classes. Bitcoin, as a store of value, maintained relatively stable futures positioning, while Ethereum, as a smart contract platform, saw its futures positioning more susceptible to application-layer volatility. A notable variable is that this quarter-end stability did not persist; by end-July, futures open interest rebounded to approximately $114 billion, with both Bitcoin (around $48 billion) and Ethereum ($25.74 billion) recovering from Q2 lows. This rebound suggests the market rebuilt bullish momentum after the brief deleveraging adjustment. However, open interest levels do not equate to absolute leverage totals, as some positions are hedged against spot longs to achieve delta-neutral exposure, meaning OI alone cannot directly indicate overall market leverage. Therefore, interpreting futures data requires integrating other indicators. From a trend perspective, the futures market's deleveraging also exhibits controlled decline characteristics rather than passive forced deleveraging, providing room for subsequent market recovery.
Key highlights show that overall, crypto-collateralized lending contracted by $11.33 billion (-16.78%) in Q2 2026, falling to $56.16 billion, down 40.13% from the Q3 2025 peak of $78.69 billion. This significant contraction primarily stems from DeFi lending applications' dollar-denominated outstanding loans declining for the third consecutive quarter, decreasing $7.79 billion (-27.61%) to $20.43 billion. DeFi lending's quarter-over-quarter contraction exceeded CeFi's, causing DeFi's former scale advantage to disappear. By end-Q2 2026, DeFi lending applications' market share fell to 47.05%, down 555 basis points quarter-over-quarter; at end-Q1, this share was 52.6%. This share reversal marks a profound shift in the power structure of the crypto lending market. A more critical variable is that CeFi outstanding loans exceeded DeFi lending applications for the first time since Q3 2023. This phenomenon suggests institutional investors are regaining trust in centralized lending platforms, or retail investors increasingly prefer CeFi services with more robust regulation. Either way, it signals the crypto lending market entering a new phase.
From the data, DeFi lending applications' outstanding borrowings have been shrinking since reaching an all-time high of $47.13 billion on September 19, 2025; as of July 21, 2026, the figure stands at $21.94 billion, down $25.19 billion or 53.45% from the peak. Since end-Q1 2026, the pace of DeFi borrowing decline has intensified, though recent signs show slight moderation, potentially indicating the DeFi lending market is approaching a bottom. On the centralized finance (CeFi) front, as of June 30, Galaxy Research tracked CeFi outstanding borrowings at $22.98 billion, down 9.62% quarter-over-quarter (a decrease of $2.45 billion). Compared to the bear market low of $6.8 billion in Q4 2023, this represents growth of $16.14 billion or 235.94%; however, it remains 37.16% below the historical peak of $36.58 billion from Q1 2022. The Q2 contraction in total CeFi lending was primarily driven by a decline in Tether-secured outstanding loans; Galaxy, Coinbase (COIN.US), Ledn, Arch, Sygnum, and Milo all saw their lending books grow during the quarter. Tether remains the dominant force in the CeFi lending market with a 58.54% market share, down 371 basis points quarter-over-quarter. Together with Maple (8.91%, +52bp QoQ) and Nexo (7.51%, +49bp QoQ), the top three institutions account for 74.96% of the market, with combined share declining 270 basis points quarter-over-quarter. When comparing market shares, it's important to note significant differences among CeFi institutions: some only offer specific loan types (e.g., BTC-collateralized only, altcoin-collateralized only, or fiat rather than stablecoin cash loans); some serve only specific client segments (institutional/retail); and operations are subject to jurisdictional restrictions. These factors result in varying expansion capabilities across institutions. Notably, CeFi data collection is challenging: accounting standards for outstanding loans differ across institutions, disclosure frequencies vary, and comprehensive data sources are difficult to obtain. Data provided by private third-party providers has not been formally verified by Galaxy Research, warranting caution in analysis. Structurally, CeFi market concentration is increasing, with top institutions' advantages becoming more pronounced while smaller players face greater competitive pressure.
Combining DeFi applications and CeFi lending platforms, total crypto-collateralized outstanding borrowings at quarter-end stood at $43.41 billion, down $10.24 billion (-19.08%) quarter-over-quarter, with the contraction primarily coming from on-chain borrowing. Worth noting: this marks the first time since Q3 2023 that CeFi outstanding loans exceeded DeFi lending applications. Note: There is a risk of double-counting between CeFi ledger totals and DeFi borrowing statistics. Some CeFi institutions lend to off-chain clients via DeFi protocols. For example: a CeFi institution stakes idle BTC, borrows USDC on-chain, then lends the USDC to off-chain borrowers. This loan would be counted in both DeFi outstanding borrowings and the institution's financial statements as a client loan. Due to lack of disclosure and on-chain identity tagging, filtering out such double-counting is difficult. DeFi lending's quarter-over-quarter contraction exceeded CeFi's, eliminating DeFi's former scale advantage. By end-Q2 2026, DeFi lending applications' market share fell to 47.05%, down 555 basis points quarter-over-quarter; at end-Q1, this share was 52.6%.
The third major segment, the crypto-collateralized portion of collateralized debt position (CDP) stablecoins, declined $1.09 billion (-7.86%) quarter-over-quarter. The same double-counting risk applies: some CeFi institutions raise funds by minting CDP stablecoins and then lend to off-chain clients. Overall crypto-collateralized lending contracted $11.33 billion (-16.78%) in Q2, reaching $56.16 billion, down 40.13% from the Q3 2025 peak of $78.69 billion. End-Q2 2026 segment market shares: DeFi lending applications: 36.37% (-544bp QoQ); CeFi lending platforms: 40.93% (+324bp QoQ); CDP stablecoin crypto-collateralized portion: 22.7% (+220bp QoQ). Combining DeFi lending and CDP stablecoins as the on-chain lending track, combined market share was 59.07%, down 324 basis points quarter-over-quarter. This data indicates that while on-chain lending's overall share is declining, its internal structure is optimizing, with CDP's steady growth providing liquidity support to the on-chain ecosystem.
In the interest rate market, using a 7-day moving average basis, from March 31 to June 30, the weighted average borrowing rate for stablecoins rose 27 basis points during the quarter; after quarter-end, rates continued higher to 3.88%. This indicator combines borrowing costs from lending protocols and CDP stablecoin minting fees, weighted by outstanding borrowing scale. The chart below breaks down two cost categories: borrowing stablecoins via lending protocols and minting CDP stablecoins with crypto collateral. Both rate trends are highly correlated, but CDP minting rates show less volatility, as these rates are manually set periodically and do not track market movements in real time. Over the past 21-plus months, both rates have been supported by the U.S. federal funds rate as a floor. In Q2, the USDC over-the-counter (OTC) benchmark rate ranged between 4.25%-5%; at quarter-end it held at 4.25%, a level maintained through August 3. USDT OTC lending rates also fluctuated in the 4.25%-5% range. For Bitcoin, the chart displays weighted borrowing rates for wrapped Bitcoin (WBTC) across multiple public chain lending applications. On-chain WBTC is primarily used as collateral, with limited borrowing demand, keeping borrowing costs persistently low. Unlike stablecoins, BTC on-chain lending rates are very stable, with low borrowing and repayment frequency. In Q2, on-chain BTC lending rates fluctuated in the 0.44%-0.5% range. The historical spread between BTC on-chain and off-chain (OTC) lending rates persisted during the quarter. OTC market BTC borrowing demand stems from two sources: ① shorting Bitcoin; ② using BTC as collateral to borrow stablecoins or fiat. Short-driven demand is uncommon in the on-chain lending market, creating the on-chain versus off-chain cost differential. BTC OTC rates held at 1% throughout the quarter.
For ETH and stETH, the chart below shows weighted borrowing rates for ETH and stETH (staked Ethereum generated via Lido protocol) across various lending protocols and public chains. Historically, ETH borrowing costs have exceeded stETH's because ETH borrowing demand is stronger. Users frequently borrow ETH to execute loop leverage strategies: staking stETH (the receipt from staking ETH via Lido), borrowing ETH, and gaining leveraged exposure to Ethereum staking annualized yields. In normal market conditions, ETH borrowing costs average within 50 basis points above or below the Ethereum staking APY. Once borrowing costs exceed staking yields, the strategy becomes unprofitable, so borrowing APY cannot sustainably exceed staking yields for extended periods. Similar to WBTC, stETH is mostly used as collateral, so borrowing stETH costs are typically low. Users can use yield-bearing liquid staking tokens (LSTs) and liquid restaking tokens (LRTs) as collateral to borrow ETH at very low or even negative net rates. This gives rise to the classic loop strategy: repeatedly depositing LST/LRT as collateral, borrowing unstaked ETH, staking that ETH to obtain new LST/LRT, then borrowing more ETH to amplify staking yield exposure. The strategy's viability depends on: ETH borrowing costs being lower than the staking annualized yield achievable from LST/LRT. Except for a few exceptional periods, this strategy generally operates smoothly. ETH OTC lending rates resemble BTC's: on-chain borrowing costs are typically lower than OTC lending. Two main reasons: short-seller-driven off-chain borrowing demand is rare on-chain; and Ethereum staking yields serve as a floor for OTC lending rates, as asset suppliers are unwilling to lend off-chain at rates below staking yields. Conversely, in on-chain markets, staking yields often act as a ceiling for ETH lending rates.
At the Aave V3 micro level, the following provides a filtered deep-dive analysis of the Aave V3 core instance (currently the largest on-chain lending market), using three screening rules: Minimum debt threshold ($100): positions below this threshold are excluded from aggregate statistics, filtering out tiny dust positions but biasing results toward larger loans. Health factor (HF) reporting cap (HF ≤ 50): positions with health factors above 50 in the snapshot are excluded from core statistics; such overcollateralized positions are typically small and of limited risk analysis value. Within thresholds, debt-weighted health factor means and percentiles include only positions with 1 ≤ HF ≤ 50; positions with HF < 1 are excluded from health factor statistics, while HF = 1 positions are included. Higher health factors indicate safer positions; health factors below 1 mean positions have triggered liquidation conditions. Health factor formula = (Total collateral value × weighted average liquidation threshold) ÷ Total borrowings. Debt-to-equity ratio (D/E): only positions where collateral value > debt (positive net equity) are counted; negative net equity positions are excluded from D/E distribution and debt-weighted average D/E. This rule is independent of the $100 debt threshold: even if a loan exceeds $100, without positive net equity, it is excluded from D/E statistics. Based on the August 7, 2026 snapshot, 19,073 valid outstanding loans remained after filtering. "Efficiency mode (e-mode)" loans account for only 8.91% of total positions, but outstanding debt and regular loans are nearly split 50/50. E-mode is characterized by high price correlation between borrowed and collateral assets (e.g., ETH collateral borrowing WETH). In Galaxy's previous April 22 tally, e-mode debt share was near 60:40; the decline stems from reduced e-mode outstanding debt. The table below shows debt-weighted risk metrics for the filtered book, split into all positions, e-mode, and regular mode. E-mode borrowers exhibit extremely high leverage: debt-weighted loan-to-value (LTV) around 90%, debt-weighted health factor of only 1.06, and debt-to-equity (D/E) around 10.7, meaning even small collateral price shocks would push numerous positions into stress territory. Regular mode loans have much thicker safety cushions: debt-weighted LTV around 49%, health factor 1.79, and D/E around 1.07, hedging against risks where collateral and borrowed assets are uncorrelated, such as cbBTC collateral borrowing USDC. The debt-weighted D/E formula (used only for positions where Ci > Di, i.e., collateral > debt): D/E = Σi (Di × (Di ÷ (Ci − Di))) ÷ Σi Di, where Di represents individual position debt and Ci represents individual position collateral value.
Next, the composition of collateral asset dollar values in the Aave V3 core market is tallied. ETH-based collateral dominates: WETH accounts for approximately 24%, weETH (Etherfi's restaked ETH wrapper) 16%, and wstETH (Lido's stETH wrapper) 14%, with these three combined representing 54.6% of all available collateral; WBTC accounts for roughly 14%. A handful of assets carry the majority of collateral value in the book, with the remainder being stablecoins and other yield-bearing tokens. On the borrowing side: WETH represents slightly over 37% of total liabilities, a typical result of the prevalence of ETH-collateralized loop leverage strategies. Stablecoin borrowings are equally substantial: USDT around 28% and USDC around 22%, together accounting for half of total borrowings, with other currencies having smaller shares. Compared to the previous tally, WETH's share of outstanding liabilities has declined significantly from 51.1%, consistent with the aforementioned contraction in e-mode debt.
From an e-mode sub-segment perspective: e-mode collateral is highly concentrated in ETH staking/restaking wrapper tokens, with weETH alone representing 42% of this category's collateral; combined with rsETH and wstETH, these three assets account for 66.2% of e-mode collateral. This indicates e-mode risk is not diversified collateral but essentially a concentrated bet on Ethereum staking fundamentals. On the borrowing side, e-mode debt is predominantly WETH-denominated, with WETH alone representing 73% of e-mode debt, perfectly matching user behavior of using ETH-based collateral to loop-borrow ETH. Stablecoins still hold a certain share, with USDT, USDe, and USDC collectively representing a low-teens percentage of e-mode borrowings. The table below tallies e-mode positions by collateral asset, including debt-weighted risk metrics; for the sub-sample where collateral is 99% concentrated in a single asset, implied loop leverage counts are also calculated. This serves as a "which asset gets loop-leveraged most aggressively" ranking, distinct from standard market capitalization ordering tables. Liquid staking and restaking ETH wrappers top the list, with high debt-weighted LTV, D/E generally in the high single digits to teens, health factors only slightly above 1, corresponding to numerous tightly looped ETH leveraged positions. The implied loop count formula for individual positions: N_i = ln((1 − (D/E)_i(1 − Li)) / Li) / ln(Li), where Li is the position's LTV; applied only to the sub-sample with ≥99% single-asset collateral. Output is the debt-weighted average Ni; positions with L outside (0,1), non-positive log arguments, or non-finite results are excluded. Values are output only for liquid staking ETH, liquid restaking ETH, yield-bearing stablecoins, and Pendle PT tokens.
Following the significant futures market decline of October 10, 2025, Q2 2026 further confirms that crypto market leverage is being gradually digested. The lending market is descending the stairs rather than taking the elevator down, with three consecutive quarters of moderate contraction instead of the single-quarter cliff crash seen in the 2022 bear market. Corporate treasury debt and futures open interest similarly reflect controlled declines rather than passive forced deleveraging. Early July data already suggests futures open interest and DeFi borrowing may be approaching bottom territory. Industry-wide crypto-related debt stands at $73.2 billion, down 15.08% quarter-over-quarter, marking the third consecutive quarterly decline. Should this trend continue, the market possesses greater capacity to absorb further contraction, potentially avoiding the cascading liquidations and counterparty collapses of the previous cycle. For now, deleveraging continues to progress step by step. This marks the first time since the 2022 bear market that the crypto market has demonstrated such resilience and self-regulatory capacity, foreshadowing a more mature and stable phase of market development ahead.
Comments