
Crypto lending protocols connect suppliers and borrowers through smart contracts. Learn how collateral, variable interest, liquidations and protocol risks work.
In brief
- Lending pools let suppliers deposit crypto assets that other users can borrow under smart contract rules.
- Most DeFi loans use collateral worth more than the borrowed amount, with liquidation possible if its value falls.
- Interest rates respond to borrowing demand, while software, oracle and liquidity risks remain.
What a DeFi lending pool does
A crypto lending protocol is a set of smart contracts that allows assets to move between suppliers and borrowers according to published rules.
Aave is an example of this model.
A supplier deposits an eligible asset into a market, and the protocol records a claim on the supplied amount.
Borrowers access available assets by providing collateral and accepting repayment obligations.
Rather than matching each loan to one named lender, many protocols pool deposits so borrowing and repayment take place against shared liquidity.
A lending market may accept assets such as Ether
ETH or USD Coin, but the supported assets and parameters depend on the particular deployment.
Suppliers receive an interest-bearing accounting position or receipt token, depending on the design.
That position tracks their claim on the pool, not a promise that the original tokens remain untouched in a separate account.
Borrowers receive tokens into their own addresses, while the smart contract tracks debt and collateral.
Rates and balances can change as the market operates.
This structure is different from a traditional bank loan.
There may be no loan officer, credit application or collection agency deciding whether an anonymous wallet can borrow.
Instead, contract checks limit the amount available against deposited collateral.
On-chain transactions record deposits, withdrawals, borrowing and repayments.
Governance or designated administrators may be able to change risk settings within the protocol's rules.
The transparency of code and transactions does not mean every borrower, dependency or economic consequence is easy to evaluate.
Collateral and borrowing capacity
Many permissionless DeFi lending markets use overcollateralization.
A borrower deposits assets whose assessed market value is greater than the amount they intend to borrow.
The excess protects the pool against ordinary price changes but does not eliminate risk.
A protocol may calculate borrowing capacity using asset-specific loan-to-value ratios, collateral factors and other limits.
Because these parameters vary, two assets with equal dollar values may support different loan amounts, and some accepted deposits may not qualify as collateral at all.
Suppose a borrower locks an asset worth more than their stablecoin loan.
If that collateral loses value, the same debt becomes large relative to the remaining collateral.
A rising debt balance from interest can produce a similar effect even without a large price move.
Protocol interfaces often express this relationship through a health factor or comparable metric.
The exact calculation follows the market's risk settings and the prices reported by its oracles.
It is not simply a comparison with the collateral's purchase price.
Collateral generally cannot be withdrawn freely while it secures outstanding debt.
The smart contract checks whether a withdrawal would leave the account inside permitted risk limits.
Some designs isolate assets or limit correlated exposures to contain losses.
When market prices change quickly, an account can move from a healthy position to one eligible for liquidation.
This creates a direct connection between the lending system and the wider market, because collateral valuation depends on asset prices outside the lending contract.
Interest rates and liquidation rules
Borrowing interest compensates suppliers for making capital available.
In utilization-based markets, the proportion of supplied assets already borrowed influences rates.
As utilization rises, the borrow rate often increases, sometimes sharply above a defined threshold.
The protocol may also allocate part of interest to a reserve or treasury.
Supply rates therefore need not equal borrowing rates.
A displayed annual percentage is a way of expressing a changing rate, not a fixed return guaranteed for the entire life of a deposit.
When collateral falls below the required safety threshold, an eligible liquidator can repay some of a borrower's debt and receive collateral under the contract's liquidation rules.
The incentive is often a discount or bonus on collateral received.
The exact amount of debt that can be repaid depends on the protocol and account condition.
Liquidation is an automated protection mechanism for lenders, but it can impose losses on borrowers.
During extreme moves, delays or insufficient market liquidity may leave the system with bad debt.
Price oracles are central to this process.
An oracle supplies values that contracts use when assessing debt and collateral.
If a price is stale, manipulated or wrong for the asset being valued, a borrower can become liquidatable too early, too late or under unfair terms.
Different protocols use different safeguards, including multiple data sources and asset-specific restrictions.
Oracle design therefore matters as much as the lending contract itself.
A pool can function exactly as programmed and still suffer when its price inputs fail.
Where the risks come from
Lending also depends on liquidity inside and outside the pool.
If most available deposits are borrowed, suppliers may be unable to withdraw the full desired amount until borrowers repay or new liquidity arrives.
This is a liquidity constraint rather than automatic evidence that deposits have disappeared.
Assets held as collateral may also trade in thin markets during liquidations.
A screen showing a large total of deposits does not establish that every supplier can exit at once without affecting prices or available balances.
Smart contract defects, compromised controls, unexpected token behavior and cross-protocol dependencies introduce other risks.
A contract audit examines selected code and assumptions, but it does not guarantee that all vulnerabilities or economic attacks have been removed.
Some lending positions are used as collateral in other applications, spreading a single failure across connected systems.
Stablecoin depegging can also change collateral quality or repayment costs.
These risks can affect suppliers even when they never borrow against their own deposits.
A useful way to understand a lending market is to separate four questions: what secures the loans, how much liquidity is available, who determines the risk parameters and how prices enter the system.
Interest comes from borrowers and sometimes additional incentives, while repayment and liquidation follow contract rules.
Aave and similar protocols make lending accessible through programmable markets, but the absence of a traditional intermediary does not remove credit, market, liquidity or technical risk.