Under the Hood: Lending 101

THORChain Lending design is revolutionary, with zero expiry, zero liquidity and zero interest (although with slippage when opening or closing loans).

This beget many questions such as:

1. Who provides the assets which are borrowed?
2. Who keeps the collateral?
3. What happens when collateral value drops?
4. Who is the counterparty?
5. What is Collateralization Ratio?
6. What are the risks?
7. Any other considerations?

It’s helpful to forget any presumptions especially from a traditional finance point of view, and try to think through the design from a first principles perspective.

Here’s an example of a loan opening and closing.

• Below examples neglect slippage and inbound/outbound gas fees
• Collateral is L1 native assets only, e.g. BTC, ETH
• Debt issued & repayment can be in any assets on THORChain, but accounted in USD value, based on internal TOR accounting unit

Loan Opening

Assuming:

• Collateralization Ratio (CR) = 200%
• 1 BTC = $20k
• 1 RUNE = $2

Loan Closing

Assuming:

• 1 BTC = $100k
• 1 RUNE = $20

From the above schematics, let’s answer some questions:

1. Who provides the assets which are borrowed?

RUNE is swapped to the borrowed asset via the liquidity pools, then sent to the borrower. Therefore, as long as the pool is available, there are always assets available to be borrowed. The deeper the pool depth, the less slippage incurred.

Remember that the pools are somewhat independent to lending. As long as the pools are healthy and functioning, any assets which are withdrawn via lending will just lead to arbers rebalancing the pools as usual.

2. Who keeps the collateral?

The collateral is swapped to RUNE via the pools. So, nobody “keeps” the collateral. Again, as long as the pools are healthy and functioning, any collateral which is deposited will just lead to arbers rebalancing the pools as usual.

3. What happens when collateral value drops?

As per above, nobody “keeps” the collateral. It has already been swapped through the pools at the moment of loan opening. So THORChain does not care, nor is under any risk if collateral value drops. This is how a zero liquidation design can be achieved.

4. Who is the counterparty?

Here is probably the crux of the whole lending design. Due to the RUNE burning/minting mechanism, it is actually all the RUNE holders which are collectively the counterparty to the borrowers. Burning/minting can be framed as concentrating/diluting the value among the rest of the RUNE holders.

Opening loans burn RUNE, and thus accrues value to all the RUNE holders. Conversely, the same RUNE holders are liable/exposed to minting of RUNE when loans are closed.

Another way to frame this is that the whole THORChain protocol is underwriting the loans. As long as RUNE has value, the lending mechanics can work as intended.

5. What is Collateralization Ratio?

The Collateralization Ratio (CR) is the ratio of collateral-to-debt (or the inverse of Loan-to-Value, LTV). A CR of 200% means user needs to deposit 200% worth of collateral when opening a loan. For example, user who deposit one BTC worth $20,000 will receive $10,000 of borrowed debt.

Unlike other lending protocols, users do not need to choose their CR when opening loans. Since there is no liquidation in THORChain lending (thus, no need to maintain the loan health above an arbitrary CR), users will automatically receive the current CR at the moment of loan opening.

CR of each pool is linearly calculated from the lending cap of each individual pool, as per https://docs.thorchain.org/thorchain-finance/lending#collateralization-ratio. Minimum CR and maximum CR are Mimir controlled, and were set at 200% and 500% respectively, at launch.

Therefore, the current CR is also an indicator for how much lending space is still available per pool. CR starts at 200% when no loans have been opened yet (or when there is max spending space), and linearly scales to 500% when the lending cap is reached. Essentially, the early borrowers generally will obtain a lower (better) CR. CR can also decrease as loans are closed, or RUNE:collateral price ratio increases.

6. What are the risks?

This is the most hotly debated aspect and it is not the intention of this article to give an exhaustive treatise. The obvious risk is excessive minting of RUNE from total loan closures vs loan openings. Factors such as the CR will strongly influence the mint/burn maths.

For a single loan opening/closing cycle, yes, nett minting of RUNE may occur, if the RUNE:collateral price ratio decreases between loan opening vs loan closure. However, it is not single loans that THORChain is concerned about, but the overall status of RUNE mint/burnt.

It could also happen that nett minting of RUNE occurs if ALL loans were closed at the same time (though there is a minimum 30 day loan period). The question is: will ALL borrowers close their loans at the same time? Some could have borrowed for actual expense needs (thus no ready cash to immediately repay). Also when some loans are closed, CR drops, and this will attract fresh borrowers.

So the idea is that NETT of [all loan closings (which would always be less than openings) vs all openings, at the RUNE:collateral price of the times of each individual openings/closings] is likely nett burning RUNE.

And even if not, there is a circuit breaker design to compartmentalize most of the lending risk, if minting causes total supply to exceed 500m RUNE. Here, the Reserves will step in to redeem the loans (instead of further minting), the whole lending design will be halted and sunset-ed, but the other aspects of THORChain will just continue as usual.

The absolute worst-case scenario is when even the Reserves are depleted trying to repay collateral. In this case, the whole THORChain will be halted (including swaps, savers, etc.), giving time for Devs and community to regroup for path forward.

Other factors which moderate these risks are the lending cap and the expected positive price impact with the RUNE burning mechanism when loans are opened. For a deeper analysis, please check out Block Science Risk Report linked below.

7. Other Considerations/Details

• When a loan is opened, the full debt (based on the current collaterization ratio) will be issued. If price of collateral increased, there is no option to request additional debt, unless user repay the original debt, redeem the collateral fully, then reopen a fresh loan.
• The same address (with an existing loan position) can open a new loan by depositing fresh collateral — new debt will be issued based on the new collateral and current collaterization ratio.
• Collateral is not returned until the full debt is repaid. Partial debt repayment can be made, but no partial collateral will be returned. Is the same address opened multiple loans, the totality of all the debt must be repaid, before the full collateral is then returned.
• Minimum loan period is 30 days (Mimir controlled).

Resources:

• Intro/summary video: Lending Essentials, Workings & Risks: https://youtu.be/AaqHG00RJks
• Lending Docs: https://docs.thorchain.org/thorchain-finance/lending
• ADR 011: https://gitlab.com/thorchain/thornode/-/blob/develop/docs/architecture/adr-011-lending.md
• Block Science Risk Report: https://hackmd.io/@blockscience/H1Q-erh_n/%2F_Btk5G8-Tn27RJfbdgecyQ

Feel free to hop into the LP University Discord to chat about this, or any other THORChain questions that you may have.

Explore THORChain: Website, X, Telegram, Developer Discord.

Explore Maya Protocol, the first friendly fork of THORChain: Website, X, Discord, Telegram.

Decentralized, permissionless, non-custodial, trust-minimized, open-sourced, economic-secured, non-wrapped, native-to-native cross-chain swaps, savings and now, lending!