Why lending protocols matter


Lending in web3 is more than just a fancy idea. With some $20bn in TVL, lending protocols have already proven to be part of the fabric of DeFi. Lending protocols have a variety of use-cases for different types of users, but we roughly divide them in four categories: users who need liquidity, users who want to speculate/hedge, users who want to optimize their capital, and users who want to earn passive yield.

The following article only discusses use-cases for over-collateralized lending protocols (the value of the assets put as collateral exceeds the total value of the loan). This sole focus on over-collateralization is not because we don’t believe in under-collateralized DeFi lending protocols (the exact opposite is even true, under-collateralized lending is a must have for DeFi to have a truly value adding impact on the world), but because not all building blocks exist within the broader DeFi ecosystem yet (maybe more on that in a later article!).

Users in need of liquidity

With lending protocols, users are not obliged to sell their digital assets for fiat money when they have immediate expenses (bills, cars, houses, starting a business…), instead they can repay the loan at a later point in time. Users might chose lending against their assets over selling them for various reasons (following list is non exhaustive):

  • By borrowing against the assets instead of selling the assets, users might avoid triggering taxable events.
  • Users can keep exposure to their digital assets when they expect prices to appreciate over time.
  • DAO’s paying out employees do not have to dump their tokens on the open market, avoiding constant selling pressure and possible reputation damage (who doesn’t like to blame the devs for dumping on their community??).
  • Users who want to retain ownership of specific digital assets, which cannot easily be bought back later in time (this one is more applicable to NFTs, where crypto natives build their identity or brand around specific digital assets eg. @punk6529).
  • Market makers (and other liquidity providers) who have a lot of capital locked in liquidity positions (eg. Uniswap V2 or V3 positions) can borrow against their existing positions instead of pulling out liquidity.

Even for users who want to sell assets because they need liquidity, lending protocols can be beneficial. The seller of an illiquid asset, for which there is a clear minimum value, can already get a cash advance on the sale while they are looking for buyers. When the item finally gets sold, the debt is automatically repaid and the seller receives the remaining proceeds. A good example is someone who wants to sell a rare NFT for which there is a liquid price floor, like a rare CryptoPunk or Bored Ape.

Users who want to speculate/hedge

Lending protocols allow (”sophisticated”) users to build non-custodial leveraged positions to speculate on directional price moves.

Speculate on increasing prices

Let’s go over an example to clarify how this would work. For this example we assume that the current price of Ethereum equals $2000 per ETH, that our user owns 10 ETH and he/she wants to build a leveraged long ETH position:

  • Our user deposits his 10 ETH in the lending protocol.
    (10 ETH → net worth $20 000)
  • He/she takes out a loan of $10 000 backed by his/her ETH position.
    (10 ETH + $10 000 - $10 000 debt → Net worth $20 000)
  • Next our user swaps his loaned $10 000 to ETH.
    (15 ETH - $10 000 debt → Net worth $20 000)
  • → If ETH now doubles in price to 4000 USD/ETH :
    15 ETH - $10 000 debt → Net worth $50 000 , compared to $40 000 if he/she didn’t use leverage.
  • → If the user is wrong and ETH drops to 1000 USD/ETH:
    15 ETH - $10 000 debt → Net worth $5 000, compared to $10 000 if he/she didn’t use leverage.

In other words, you amplify your gains and amplify your losses.

Speculate on decreasing prices

In a similar fashion, users can also take short positions against digital assets (a short ETH position in the next example):

  • Deposit USD.
    (20 000 USD → net worth $20 000)
  • Take out loan in ETH against USD position.
    ($20 000 + 5 ETH - 5 ETH debt → net worth $20 000)
  • Swap ETH to USD.
    ($30 000 - 5 ETH debt → net worth $20 000)
  • → If ETH decreases in value to 1000 USD/ETH, your net worth denominated in USD increases: $30 000 - 5 ETH debt → net worth $25 000.
  • → And opposite, if ETH increases in value to 4000 USD/ETH, your net worth decreases: $30 000 - 5 ETH debt → net worth $10 000.

Hedge against price volatility

While less commonly used, lending protocols can similarly be used to hedge against price volatility (for the real apes out there who don’t know, hedging means reducing risks).

The strategy is similar, but instead of amplifying gains and losses with price movements, you want to dampen them:

  • Deposit ETH.
    (10 ETH → net worth $20 000)
  • Take out loan in ETH against ETH position.
    (15 ETH — 5 ETH debt → net worth: $20 000)
  • Swap loaned ETH to USD.
    (10 ETH + $10 000–5 ETH debt → net worth: $20 000)
  • → When the ETH price drops 50%, your net worth drops less than 50%: 10 ETH + $10 000 - 5ETH debt → net worth $15 000.
  • → When the ETH price increases 100%, your net worth increases with less than 100%: 10 ETH + $10 000 - 5ETH debt → net worth $30 000.

In other words, you dampen the impact of price fluctuation

You might wonder: “Why go through the hassle of hedging using lending protocols, while you can simply directly sell a part of your ETH portfolio to USD???” This would indeed achieve the exact same results as hedging via a loan, but while direct selling of ETH to USD triggers a taxable event in most jurisdictions, hedging via loans might not.

Users who want to optimize their capital using sophisticated strategies

Even more “““sophisticated””” DeFi users can truly unlock the power of their capital with complex strategies, involving lending protocols and multiple other DeFi protocols.


  • Lend against yield bearing assets (for instance Yearn vaults or LP-positions in DEXes) and as long as the yield is higher than the interest of the loan, your collateral itself is repaying your loan!
  • Other possibilities are leveraged yield farming (lend against a yield bearing asset, and use the loan to increase your position in the yield farm), which boils down to leveraging the delta between the yield of the farm and the interest rate of the loan,
  • Building delta neutral strategies,
  • And many other strategies, which deserve their own separate article. (see for instance Marc Zeller and his #FrenchCharts)

Earning yield

So far we only discussed why borrowers would use lending protocols. Of course there is also the other side of the equation: the lenders, those who provide the liquidity.

The incentive for lenders to use a lending protocol is luckily a lot more straightforward: lenders want to earn yield on their existing capital.


While lending protocols are useful tools for many use-cases, it would be unfair not to mention the associated risks.


An important metric in lending protocols is the health factor of a loan. The health factor is an indication how close the borrower is to getting liquidated. It is defined as the ratio between the value of the collateral assets backing the loan and the total value of the loan itself.

Since borrowers must maintain an over-collateralized position at all time (the total value of their assets put as collateral must always be greater than their open debt position), the health factor of a loan must always be greater than 1.

The health factor can decrease in two ways:

  • The assets backing the loan decrease in value,
  • The debt increases over time, since interests are continuously compounded.

When the health factor of the loan approaches 1 and drops below a predefined threshold (originally named the “liquidation threshold”), the borrower gets liquidated: he/she looses ownership of the assets that were put up as collateral. The liquidated assets are auctioned, sold on the open market or transferred to the lenders to repay the remaining debt.


Under normal circumstances, the liquidation mechanism guarantees that unhealthy loans are liquidated before the health factor drops below 1. However during extreme events (50% crashes or extreme gas prices), it is possible that the health factor drops below 1 and the loan becomes under-collateralized. At this point in time, not all debt is backed by actual collateral.

In most cases the remaining collateral will still be sold/auctioned, and who ends up paying for the bad debt differs from protocol to protocol:

  • The lenders are not fully paid back,
  • The protocol itself has an emergency fund,
  • The equity/token holders of the protocol are diluted,
  • Third parties or underwriters pay for the difference.

In any case, under-collateralization of an over-collateralized lending protocol is a situation you’d want to avoid at all costs!

Smart contract Risks

Like any software code, smart contracts can contain bugs that can be exploited to steal user funds. This is a risk for both the lenders and the borrowers, and harder to quantify as compared the two earlier mentioned risks.

Not only the code of the lending protocol itself is vulnerable, but also third party integrations. Think pricing issues that occurred during the UST/LUNA downfall or the CREAM hack.

When designing a lending application, a lot can be learned from previous exploits and failed protocols. That alone will probably be enough content to cover in a future post!


By now you should have at least a basic understanding of the reasons why users would be willing to use a lending protocol, and how lending protocols provide value to the broader DeFi ecosystem.

“But, which lending protocol to use??? Should I use Maker, AAVE or am I cool enough to use Arcadia Finance??” you might think. Hold tight, next week we’ll write about the different approaches lending protocols take. What their different approaches are and -as you expected-, why users benefit when using Arcadia Finance as compared to using existing lending protocols.

Same time & place next week, see you there!