Lending markets: utilization, LTV, liquidation

How a collateralized lending market computes — and at which three points it breaks.

This lesson has had no expert review. It was written for this platform and against the evidence it cites; nobody has gone through it independently.

Learning objectives

  • You can tell utilization, LTV and health factor apart and compute them.
  • You can explain how market risk becomes credit risk.

Check your prior knowledge

Answer these for yourself before reading on. Wherever you hesitate is where this lesson pays off.

Core concept

Fact

Utilization drives the rate — and caps withdrawals

The rate in a lending market is a function of utilization: the more of the supplied capital is lent out, the more expensive the loan and the more reaches the depositor. The same number also describes how little free capital is left. A market at 95 % utilization pays well and may no longer let a large depositor out.

Fact

The health factor is the distance to forced sale

Instead of the borrower's creditworthiness, what counts is the value of their collateral. If it falls far enough that the position drops below the threshold, anyone may realize it at a discount. That is not an exception but the built-in mechanism keeping the market solvent — and it acts automatically, with no grace period and no conversation.

Risk

Bad debt arises when the mechanism acts too late

Liquidation presupposes someone buys — fast enough and at a price covering the loan. If the collateral falls faster than liquidation can happen, or the market is too thin, an uncovered claim is left behind. Who bears it is the protocol's decision: the reserve, the depositors or the token holders. That question rarely appears in the product description.

Definitions

Utilization
Share of supplied capital that is borrowed.
Health factor
Ratio of weighted collateral value to outstanding loan; below 1 the position is liquidatable.
Max LTV in the glossary
Highest share of collateral value that may be borrowed against.

Model

  1. Collateral price falls

  2. Oracle writes the new price into the contract

  3. Health factor drops below 1 — the position becomes liquidatable

  4. Realization at a discount, selling pressure in the market

  5. If proceeds fall short: bad debt, allocated by the protocol's rule

From a price decline to an uncovered claim — Step 4 can amplify step 1 — that is the cascade.

Formulas

Utilization

Utilization = borrowed / supplied
borrowed
sum of outstanding loans
supplied
sum of supplied capital

Limit: The figure says nothing about distribution: the same utilization can arise from many small loans or from two large ones, with very different liquidation dynamics.

Health factor

HF = (collateral value × liquidation threshold) / loan value
collateral value
current value per the oracle
liquidation threshold
a market parameter, typically below 1
loan value
outstanding loan including accrued interest

Limit: The value depends entirely on the oracle price, not the market price. Where the two diverge, the position is exposed to the oracle, not to the market.

Worked example

How far may the price fall?

Collateral
10 ETH at USD 2,000 = USD 20,000
Loan
USD 10,000
Liquidation threshold
0.80

HF = (20,000 × 0.80) / 10,000 = 1.6. HF = 1 is reached at a collateral value of 10,000 / 0.80 = USD 12,500, i.e. USD 1,250 per ETH.

Buffer to liquidation: a 37.5 % price decline.

Reading: The buffer is a price distance, not time. How long it lasts depends on the collateral's volatility — and on whether the oracle reflects the fall promptly at all.

Interactive model

The worked example, with movable figures. Estimate first what happens — then check.

Health factor and liquidation price

Move the collateral price and watch the distance to the HF = 1 threshold.

Estimate first, then check

Estimate first: at what ETH price does the example's position reach a health factor of 1? And what happens to the buffer if the liquidation threshold drops from 0.80 to 0.70?

The starting values are the worked example's own figures — change one input at a time.

Health factor: 1.6. Price at HF = 1: $1,250. above the threshold — not liquidatable.

What the model does not show: The loan value stays constant here. In practice interest accrues, and the price the model moves is the oracle price — not the market price at which a liquidation is then executed.

Retrieval

A lending market shows 96 % utilization and a markedly elevated supply rate. What follows for a large depositor?
What does a position's health factor depend on directly?

Exercise on real data

Open a lending market and note utilization and maximum LTV. Work out what collateral price decline would reach the threshold.

Look at a lending market's utilization and LTV →

Application

Which three metrics do you monitor continuously when your organization supplies capital to a lending market — and at what value do you act?

Related case studies

Institutional reading

Bank
How should the possibility of constrained withdrawals be reflected in the liquidity risk report?
Insurance
Would bad debt be an insured event or a market loss?
Asset management
At what utilization is the position reduced — and who decides that?

Metrics in this lesson

Key takeaways

Evidence