LibraryCredit and stable value2019Design paperCorpus record
Compound: The Money Market Protocol
Compound. Robert Leshner and Geoffrey Hayes.
The 2019 Compound paper: pooled lending markets where suppliers earn a floating rate and borrowers post collateral. Interest rates are a function of utilisation, set in the protocol rather than negotiated bilaterally.
Compound's paper turns deposited tokens into a fungible claim and lets borrowers post collateral and draw a balance, with interest rates that move automatically as utilisation of the pool changes.
- Evidence
- Primary paper
- Re-measured
- No
- Assumptions
- 4
- Records linked
- 2
Code
Source, repository, and what is actually hosted
- Compound white paperhttps://compound.finance/documents/Compound.Whitepaper.pdf
- The source URL is not a GitHub repository. No code remote is invented.
- Full text is not mirrored. Public availability is not a licence to copy the file onto this site.
- Add to a reading list
01 Claim ledger
What the paper is allowed to say
Each row is a sentence already in the study. The status is the same on every row: a model claim, not a live measurement.
Claim 01 · Paper model
Supply is a pool, not a matched loan
Lenders do not pick a borrower. They deposit into a market and receive cTokens. Borrowers draw against the pool if their collateral covers the debt under the protocol's factors.
Claim 02 · Paper model
Interest is a function of utilisation
As more of the pool is borrowed, the rate rises, which is meant to pull lenders in and push borrowers out. The kink in that function is a governance choice the paper makes structural.
Claim 03 · Paper model
cTokens are the receipt
The exchange rate of a cToken increases as interest accrues. A lender redeems by burning cTokens, not by recalling a specific loan.
Claim 04 · Paper model
Collateral factors decide how far a borrower may go
Each asset has a factor less than one. A borrower whose debt exceeds what the factors allow can be liquidated. The factor is the risk policy.
Claim 05 · Paper model
Liquidation is a discount, not a courtroom
A liquidator repays part of the debt and seizes collateral at a discount. The paper relies on liquidators showing up while the collateral still covers the debt.
02 Three cuts
Observation, model, falsifier
A desk does not stop at the summary. Each claim is cut three ways, using only this study's own assumptions and checks. Nothing here is a new figure.
01 Supply is a pool, not a matched loan
Observation
What the study says
Lenders do not pick a borrower. They deposit into a market and receive cTokens. Borrowers draw against the pool if their collateral covers the debt under the protocol's factors.
Model
What has to hold
Collateral factors are set below realistic crash sizes. Governance that raises them is taking risk on behalf of lenders.
02 Interest is a function of utilisation
Observation
What the study says
As more of the pool is borrowed, the rate rises, which is meant to pull lenders in and push borrowers out. The kink in that function is a governance choice the paper makes structural.
Model
What has to hold
Collateral factors are set below realistic crash sizes. Governance that raises them is taking risk on behalf of lenders.
Falsifier
What would retire it
What is the collateral factor, and which oracle feeds the health check?
03 cTokens are the receipt
Observation
What the study says
The exchange rate of a cToken increases as interest accrues. A lender redeems by burning cTokens, not by recalling a specific loan.
Model
What has to hold
The underlying token rebases or confiscates nothing. The cToken math assumes balances behave.
Falsifier
What would retire it
Where is the kink in the interest model, and who sets it?
04 Collateral factors decide how far a borrower may go
Observation
What the study says
Each asset has a factor less than one. A borrower whose debt exceeds what the factors allow can be liquidated. The factor is the risk policy.
Model
What has to hold
Collateral factors are set below realistic crash sizes. Governance that raises them is taking risk on behalf of lenders.
Falsifier
What would retire it
What is the collateral factor, and which oracle feeds the health check?
05 Liquidation is a discount, not a courtroom
Observation
What the study says
A liquidator repays part of the debt and seizes collateral at a discount. The paper relies on liquidators showing up while the collateral still covers the debt.
Model
What has to hold
Liquidators can act before collateral gaps too far. Congested blocks delay them.
Falsifier
What would retire it
What is the liquidation incentive, and is there a liquidator when the chain is congested?
03 Sequence
One action, as an operating tape
- 01A lender deposits an asset and receives cTokens at the current exchange rate.
- 02A borrower supplies collateral, then calls borrow. The protocol checks the sum of collateral values times factors against the new debt.
- 03Interest accrues each block according to the utilisation model. The borrow balance rises. The cToken exchange rate rises.
- 04If the borrower's health falls through the threshold, a liquidator repays some debt and receives collateral plus the liquidation incentive.
- 05The lender redeems cTokens for the underlying, receiving their share of the pool including interest and any shortfall the pool absorbed.
04 Load-bearing
The argument, and where a pitch drops it
The rate model is a governor, not a promise
The cut
A high borrow rate does not create a lender who is not there. If utilisation sits at the kink because liquidity left, borrowers simply pay more until they are liquidated or markets move. The paper should be read as a feedback loop with failure modes, not as a guarantee of a moderate rate.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
Later Compound versions and the COMP distribution are not this paper.
Oracle prices are inside the solvency check
The cut
Collateral factors apply to a price. If the price is stale or manipulable, the health check is fiction and liquidations are either late or unfair. The paper's risk section depends on a price the 2019 text does not fully decentralise. Name the oracle.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
A collateral factor is a risk parameter, not a scientific constant.
cToken composability spreads the pool's risk
The cut
Once cTokens are used as collateral elsewhere, a shortfall or a frozen market is no longer local. The paper's fungible receipt is what makes that possible. Usefulness and contagion are the same feature.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
A collateral factor is a risk parameter, not a scientific constant.
05 Register
What has to be true
Network · Not re-measured
Liquidators can act before collateral gaps too far. Congested blocks delay them.
Model · Not re-measured
The price feed is timely. A paper solvency check on a wrong price is insolvency.
Model · Not re-measured
Collateral factors are set below realistic crash sizes. Governance that raises them is taking risk on behalf of lenders.
Model · Not re-measured
The underlying token rebases or confiscates nothing. The cToken math assumes balances behave.
06 Divergence
What happened after the paper
Compound v2 became a reference money market, and later versions and forks changed the rate model, the oracle and the governance token. The Comptroller risk engine is the piece to study. A current supply APY is not in the paper and should not be implied by it.
A later client, parameter set, or reward formula is a different object. Cite this paper for the mechanism. Cite a primary release for the network. This desk has not re-run the proof.
07 Pre-mortem
What to check before you use the idea
- 0 of 4 marked on this browser. A mark is a reading note, not a pass, a rating, or a recommendation.
08 Anatomy
The paper, in the order a builder needs
The problem it names
Matching a lender to a borrower is slow, and a bilateral loan does not let a supplier exit without the borrower's cooperation. Compound pools suppliers of one asset and borrowers of that asset, and prices the rate from how full the pool is.
What the design proposes
- Each asset is a market. Supply and borrow balances are claims on that market, not IOUs against a named person.
- A collateral factor limits how much a borrower may take against a given asset.
- If health falls through a threshold, a liquidator may repay part of the debt and seize collateral, with a discount.
How the mechanism is specified
- The interest-rate model is a public function of utilisation. Changing the function is a governance action, which the paper locates outside the core math.
- Interest compounds into the exchange rate of the claim token, in the paper's accounting.
- Oracle prices determine collateral value. The paper needs an oracle even where it does not dwell on one.
What this page does not treat as proven
- A collateral factor is a risk parameter, not a scientific constant.
- Liquidation only works if someone is willing and able to do it, and if the oracle is timely.
- Later Compound versions and the COMP distribution are not this paper.
Why the desk still reads it
The template for on-ledger money markets. A venture copying it needs an answer on oracles, on who sets collateral factors, and on what happens when the asset cannot be liquidated at the price the oracle printed.
09 Lexicon
Terms, opened into the record
- cToken
- A receipt whose exchange rate rises as borrowers pay interest into the pool.
- Utilisation
- The fraction of a market's deposits that is currently borrowed. The interest model reads this number.
- Collateral factor
- The fraction of an asset's value that counts toward borrowing power.
- Liquidation
- A third party repaying debt to seize discounted collateral when a borrower is under the health threshold.
10 Repository
Every linked record on this page
Underlined words open a page that already exists: a concept, a protocol profile, a failure record, or another paper. If a word is not underlined, this desk does not have a record for it.
Concepts
This is Blockchain Lab's reading of a public design paper. It is not the paper, not a copy of it, and not an offer of tokens, equity, custody or a partnership. Later network behaviour can diverge from the text. Nothing here is investment, legal or technical advice.
Research status: Design paper. Last reviewed: 1 October 2026. This is a reading of a public paper, not investment, legal or security advice.