Author: Climber, CryptoPulse Labs
On April 22, on-chain data showed that Aave’s total deposits dropped from a high of $48.5 billion to $30.7 billion. Approximately $15.1 billion flowed out within a short period, with nearly one-third of the deposit volume being reallocated.
Meanwhile, the funds did not exit DeFi but were reallocated among different lending protocols. Morpho experienced an outflow of about $1.5 billion, while Spark saw a counter-trend increase of approximately $1.3 billion, absorbing funds from entities including Justin Sun and large whales who had made significant bottom-fishing purchases in February this year.
On the surface, funds flowed from Aave to Spark, but in reality, users are re-selecting more stable and credible platforms, which is also a repricing of risk credibility around on-chain lending protocols.
I. Aave Capital Outflow: A Major Retreat After Risk Exposure
On April 18, the Kelp DAO cross-chain bridge was attacked, with the attacker minting approximately 116,500 rsETH without real asset backing and depositing it into the Aave system for lending operations. This action directly triggered Aave’s risk mechanism, and Guardian subsequently urgently froze the related assets.
However, what truly shifted market expectations was not the freezing action, but the result of the reassessment of on-chain risks after the attack occurred.
On-chain analysis data shows that Aave’s overall deposits fell from about $48.5 billion to $30.7 billion within just three and a half days. Approximately $15.1 billion flowed out, close to one-third of the capital scale. Changes of this magnitude are not mere panic but a reallocation of systemic funds.
Meanwhile, the structural changes in project funds are also very clear.
High-risk asset pools were withdrawn first, particularly those related to cross-chain assets and derivative staking assets. Neutral assets remained within the Aave system. Institutional funds began splitting positions and dispersing flows to other lending protocols.
This indicates that the market’s judgment was not to leave Aave but to redefine Aave’s risk boundaries. Aave’s issues were thus amplified because its risk mechanism is event-driven.
Assets operate within the system after entry, and freezing and remediation only occur when anomalies arise. However, from an institutional capital perspective, this implies a core issue: risks are identified only after being triggered, rather than being filtered before entering the system.
Therefore, in a lending market with a scale of hundreds of billions of dollars, differences in such structures directly influence the long-term staying tendency of funds.
II. Spark’s Capital Absorption Logic: The Good Fruit of Rules First
While Aave experienced large-scale capital outflows, Spark’s performance was entirely opposite.
SparkLend TVL rose from about $1.9 billion to $3.2 billion, with a short-term increase of approximately $1.3 billion. Moreover, the source structure of funds showed clear institutional characteristics, including Justin Sun and several large whales reallocating funds into the Spark system.
But the key point is that this round of capital inflow was not because Spark performed better in this incident, but because its risk structure itself is different.
Spark is backed by the MakerDAO (Sky) ecosystem, and its core logic is not to accept assets but to screen assets.
Long before this rsETH incident occurred, Spark had already adjusted or even restricted the risk parameters of related assets at the governance level. This decision was not based on whether Kelp DAO had vulnerabilities, but on a long-term risk assessment of the collateral asset model.
This means that while Aave needed to handle risks during the incident, Spark structurally avoided risks from entering the system. In other words, risks were cut off and isolated at their source of occurrence.
Therefore, in this round of changes, Spark did not experience asset freezes, bad debt risks, or systemic shocks, so it attracted another type of funds that are more sensitive to the certainty of rules.
Especially institutional funds, which focus not on a particularly high numerical value of yield but on whether risks are already determined to be zero before entering the system.
Thus, Spark offers not higher yields but clearer capital pathways. For example, whether assets can enter is determined by rules. After entry, risk boundaries are stable. Yields are systemic structures, not market volatility outcomes.
III. Will Spark Become the Next Aave?
If only looking at capital flows, it’s easy to draw an intuitive judgment that Spark has been continuously absorbing funds flowing out of Aave. But if we broaden and enlarge our perspective, we find that the two are not even on the same competitive dimension.
Because Aave represents the first stage of DeFi lending, which is an open market mechanism.
Here, assets enter freely, and interest rates and risk premiums are determined by the market. The advantage of this model is scale and liquidity, but the problem is that risks are often discovered after the fact.
Spark represents the second stage: a governance preset mechanism.
That is, assets are screened before entering the system. Risks are identified and resolved at the rule level in advance, rather than being handled after exposure in the market.
In other words, Aave is market-driven risk, while Spark is rule-driven risk.
Additionally, from a capital perspective, this also implies a structural change: the lending market is evolving from a single liquidity pool into a tiered and hierarchical system.
Thus, Aave remains the largest and deepest general lending market, and its liquidity and asset coverage cannot be replaced in the short term. But Spark is becoming another role: a structured entry point for stablecoins and institutional funds.
The relationship between the two is not substitution but division of labor. Aave handles market liquidity, while Spark handles rule-based capital pathways. Therefore, Aave will still be the leading player in the lending market in the short term, while Spark still has a long way to go.
Conclusion
The Kelp DAO cross-chain bridge incident is merely a trigger; what it truly changed is not the deposit scale of a particular protocol, but how funds perceive the way risks are defined.
This round of changes in capital flow does not mean Aave is weakened, nor does it imply Spark will replace Aave. Rather, the lending market is transitioning from competition among single liquidity pools into a new stage of risk structure stratification.
