Clearpool
Published in

Clearpool

Clearpool x zkLend AMA Recap

On 15th August, Jakob Kronbichler (COO & co-founder of Clearpool) joined zkLend for an AMA on Twitter Spaces.

Click here to listen to the recording on Spotify.

Below is a summary of the highlights and insights that Jakob shared on Clearpool and the future of multichain lending.

What was the idea that made you guys interested in the Web3 and DeFi space and then actually coming up with the idea of building Clearpool?

Jakob: Clearpool is actually the brainchild of Rob and Alessio originally. Rob and Alessio worked together at the bank. Rob had spent his entire career in capital markets, and Alessio used to work for him. Alessio left 3–4 years ago to start a different crypto company called Hex Trust, a digital asset custodian based out of Hong Kong that also has offices in Singapore. They knew each other well and were constantly talking about DeFi when they noticed a gap in the market.

The original leading DeFi protocols, such as AAVE, Compound, and Maker, were working really well, but they were only providing over collateralized loans that weren’t particularly capital efficient and didn’t serve some of the institutions in this space that were looking for a more efficient source of capital. This is when the idea for Clearpool arose. Meanwhile, I became more involved in the crypto space and met with them in Hong Kong; with my background in lending, we thought it was a good fit. So that’s where Clearpool began last year.

We then raised a funding round in October of last year, led by Sequoia Capital, Arrington, Sino, Wintermute, and the participation of several other prominent investors. Following that, we launched the Clearpool protocol on mainnet in late March. It’s been a very turbulent five months, with a lot happening in the unsecured lending space, especially with everything that has happened around 3 Arrows Capital (3AC). To give you a sense of where we are, we currently have around US$100 million in active liquidity provided to the platform.

Other lending protocols, such as AAVE and Compound, use over collateralized approaches, whereas Clearpool takes a more unique approach and serves a more niche market. One, there is no collateral, and two, it is very defined, as it uses single-borrower pools and single currency as well. All of these are very niche markets for crypto-native funds and financial institutions alike. Was there a specific conviction when you started this, and how did the founders derive the entire unique strategy?

Jakob: First of all, we wanted to concentrate on the under-collateralized lending market. It was critical that the protocol only focuses on under-collateralized lending and not on over-collateralized lending. Then, there are two major reasons why we chose single borrower pools.

  1. Clearpool is envisioned as an unsecured liquidity marketplace. Lenders can enter the market, view the various borrower profiles, and compare them using various indicators to make a decision. Lenders are also encouraged to conduct their due diligence and research to achieve the desired risk-return profile. Clearpool aspires to be a decentralized credit marketplace where lenders have the transparency and authority to decide who they want to lend to.
  2. We started with single borrower pools because we believe they will serve as the foundation for a couple of additional products that will be built on top of them, such as thematic pools that are on the way. It will be automated bots with a mandate to invest in various particular pools. For example, a mandate to only invest in certain risk profiles, such as AA-rated borrowers.

There are also other opportunities. Lenders who provide liquidity on Clearpool will receive cpTokens that are unique to each borrower pool. As a result, we expect the development of a secondary market for cpTokens, as well as the development of additional products such as credit default swaps and others on top of these pools.

That was the reasoning behind our decision to use single borrower pools. The borrower profile is currently exclusive to trading firms, the majority of which are crypto-native trading firms. We work with some of the biggest names in the industry, including Wintermute and Amber, as well as Jane Street, one of the largest Wall Street trading firms. These trading firms use risk-neutral strategies to execute a wide range of trades and capitalize on minor price inefficiencies across multiple exchanges. This was the apparent first borrower profile for which we felt comfortable in assessing the risk and so on. However, in the future, we intend to explore different types of borrowers and currencies, including not only stablecoins but also other cryptocurrencies.

Could you please explain how borrowers can approach Clearpool to establish a borrowing pool? Is a specific risk-neutral strategy required for them to be onboarded for the initial batch, or can they have a different strategy?

Jakob: We are open to different borrower profiles in the future. However, we believe that these market risk-neutral high-frequency trading firms are ideal counterparties at the moment, and we also see that the market is still quite large. We are currently focusing on those firms rather than those that employ directional trading strategies that are difficult to monitor, particularly in today’s market.

In general, we screen the companies, interview and question them about their trading strategies, and have specific criteria that must be met before they can proceed to the onboarding process. The borrowers will then be KYC’ed and subjected to a risk assessment process conducted by our partner, Credora. They were previously known as X-margin and were recently rebranded.

Credora provides an innovative approach in addition to regular financial statement analysis and due diligence. They have the ability to perform real-time risk monitoring, and borrowers must provide API access to the various trading accounts with which they are working, as well as some default wallets. Using these approaches, Credora can calculate a real-time score based on their equity across all of the different exchanges and for example, whether they are delta-neutral or not. Borrowers are still whitelisted by the team for the time being, but this function will be performed entirely by the Community in the future. It should be noted, however, that once a borrower pool is launched on Clearpool, it is effectively empty, and it is up to the lenders to decide who they want to fund.

When all of these Credora results and metrics come in, will CPOOL token holders eventually have the governance to control who gets onboarded or some of the metrics for getting into the borrower pools?

Jakob: Currently, there is still some involvement from the team in talking with these borrowers and deciding whether to continue or not with a specific borrower. Following the assessment by Credora, we observed that certain borrowers are probably not a risk profile with which we are comfortable. However, we will gradually decentralize, and we hope to launch this governance piece very soon, where our CPOOL token holders will be able to decide whether a borrower is accepted to launch a pool on Clearpool.

More importantly, the new governance piece will also decide the interest rates. CPOOL token holders can delegate votes to Clearpool oracles, who will regularly input four different parameters for the interest rate curves. Thus, our governance piece will determine the borrowers’ payment amount. More information on delegated staking and governance will be made available shortly.

TrueFi and Maple Finance are the two main contenders right now. What is Clearpool’s unique selling point, how does it differ from other protocols, and why would users choose Clearpool over the other alternatives?

Jakob: The first major distinction is that Clearpool is a marketplace with various individual borrower pools. In Maple and TrueFi, you would fund a summary pool, and other parties would underwrite and allocate those funds. In our case, lenders visit the platform, view and compare different borrower pools, and then make their own decision on exactly which borrower pool to fund.

The second is the pool’s dynamic nature. Maple, TrueFi, and a few smaller protocols offer fixed-term loans with a fixed interest rate and term duration. Clearpool’s borrower pools are very dynamic as it does not have a fixed pool size or interest rate. They are determined by the pool’s utilization rate, which means that the pool is determined purely by market forces, such as how many lenders deposit into the pool and how much liquidity the borrower currently borrows.

Clearpool has an interesting dynamic curve rate that incentivizes borrowers by offering the lowest interest rates when utilizing roughly 85% of the liquidity. So, if a borrower uses too little liquidity, the interest rate rises to compensate lenders for the fact that they are not using it. The same was true for very high utilization rates, as interest rates would rise as well, and borrowers would be required to repay their loans to bring them down. This dynamic rate curve is intended to ensure some exit liquidity within the pools, so lenders may withdraw at any time as long as there is sufficient liquidity.

When a large number of lenders withdraw their funds in the event of a market shock, such as the LUNA, UST, and 3AC incidents, we see that the pools can scale down relatively quickly, allowing lenders to withdraw their funds if they are no longer comfortable with the risk. In contrast, when market sentiment is very positive, the pools can also scale up quite quickly.

I’ve worked in marketplaces my entire professional career, and I’ve discovered that every marketplace has two important sides. In the case of Clearpool we see a lot of demand on the borrowing side, and we identified the lending side to be the harder one to scale. This is why we chose a solution that is quite favorable for lenders, giving them full control and flexibility in the protocol. Furthermore, the recent CeFi lending incidents have caused lenders to be a little more cautious right now. They expect greater transparency and flexibility, which is why our product fits quite well in this market.

Borrowers also benefit from the flexibility offered by Clearpool. They can withdraw more liquidity at any time, and they can also repay and close the loan in place. I wouldn’t say we’re better or worse; we just have a slightly distinct product. They have fixed-term loans, whereas we have these incredibly dynamic pools.

You mentioned an optimum utilization rate of around 85%. How it’s being stabilized at that point where borrowers and lenders can meet because it’s quite interesting to see a lot of the pools just stop around that rate number.

Jakob: That’s actually a very good point. Initially, we had a different curve, with interest rates rising in tandem with the increasing utilization rate. The logic behind the initial curve rate is that if utilization and interest rates are both too low, lenders will withdraw, and a new equilibrium will be established. We realized that the pools were inefficient, so we devised this U-shaped cosine curve. It has a U-shape, starting with a higher interest rate, gradually decreasing until it reaches the optimal level (currently around 85%), and then rapidly increasing again. As a result, the new curve discourages usage in the extreme low and high ranges.

When we implemented this new curve, we observed that borrowers rebalanced the pool much more frequently, keeping it for now balanced around 84–85%. As a result, for the time being, the pools are steadily growing, and borrowers would monitor and draw a little more liquidity every day or even repay if any funds were taken out. This utilization rate of 85% with the lowest interest rates has a strong incentive to keep it as low as possible because every percentage matters to them.

Let’s now discuss risk management. You mentioned earlier that Clearpool has been integrating with Credora and using them for part of the credit underwriting work. Could you perhaps provide some more details regarding this integration and how it functions in conjunction with the Clearpool credit management design?

Jakob: It’s important to understand that there isn’t a true integration per se; rather, we use Credora as an external provider of this credit rating service to offer our lenders access to the score and borrowing capacity they need to make a well-informed decision. Credora, a supporting partner, provides the score, while Clearpool essentially provides the marketplace. Since Credora provides a really powerful solution and is at the forefront of this industry, we are excited to partner with them.

In the future, if other firms analyze different things or are better at analyzing different counterparties, we would be happy to consider showing some of those scores and inputs as well. Finally, it all comes down to providing our users with the information they need to make an informed decision. However, for the time being, borrowers should obtain this rating, which we will display on our front end.

As I mentioned previously, the cool thing about Credora is that, in addition to looking at static documents such as financials, they can see in real-time whether the underlying credit rating improves or deteriorates and if there are any real-time changes. Credora has API access to borrower accounts and can observe the actual trading that is taking place, giving them a real-time view under the hood of these firms. We realize how powerful this is in the current market situation, which is obviously a very risky situation in which many lenders are rightfully concerned and want to make decisions based on the most accurate information available.

What would be your response if someone asked you whether the user is sophisticated enough, particularly for retail users who may not have a TradFi or DeFi background or even enough knowledge, to analyze or assess these credit account party risks?

Jakob: One of the most important aspects is to be completely transparent about the risks associated with a protocol. We are in the unsecured lending space, and every protocol participant must understand that there is some risk involved. If an unsecured lending protocol grows to a certain size, it will face a default at some point. That’s why lenders are compensated with higher risk-adjusted rates than over-collateralized protocols like Compound and AAVE.

That’s why we went down this route, to provide a marketplace that will attract a large number of sophisticated investors and institutions that may be familiar with the counterparties and have done some due diligence on their own, even without the credit scores provided. The collaboration with Credora is also important here. There is detailed documentation that explains how the score is computed in detail so you can see which factors provide what weight and why a particular rating is an outcome.

In fact, we want to be even more specific, so instead of just using letters like A, AA, or BB, we want to include the many areas such as financials, due diligence, risk monitoring, and so on. We would like a separate section for each of those various categories to clarify what kind of information has been provided. We want to be as clear as possible in order to give each participant the best consideration in deciding whether or not to join the pool.

One final question about risk management. Would there be any forewarning or update in Clearpool’s uncollateralized setup, for example, decreasing Credora rating metrics, and how would that translate to Clearpool taking action as the user’s first line of defense?

Jakob: First, if a borrower provides all required information, the Credora rating will be real-time. So, if there is a significant drop in their assets, for example, that should theoretically be recorded and result in a lower rating. Second, there’s the pool utilization feature. If a pool’s utilization rate exceeds 99% for more than five calendar days and the borrower fails to repay to reduce it to at least 95%, a default scenario is triggered.

In the event of a default, the insurance pool that accumulates over time will then be deployed and made available to all LPs. Borrowers were also required to sign a set of legal documents, which included a lending agreement and stated that legal recourse was available in the event of a default.

When a default occurs, an auction will be automatically triggered in which participants can bid on the pool’s cpTokens. Following the auction, we are implementing a voting process in which all cpToken holders will have the opportunity to accept or reject the winning bid. If the majority of cpToken holders agree, they can redeem their cpTokens for their proportionate share of the winning bid amount and choose not to pursue the borrower. However, if they do not accept, they have the legal right to pursue the defaulted borrower individually. Then again, it is critical to understand that there is risk involved, and there will be losses in default cases.

Over the last few months, events such as LUNA, UST, and 3AC have caused a market shock for a number of protocols across the sector, and overall TVL across the majority of DeFi lending protocols has decreased significantly as most investors derisk. What measures do you believe we should take during this time period to increase user confidence and overall protocol functionality for long-term success?

Jakob: Transparency is key. Personally, I believe that algorithmic stablecoins like UST and lending protocols like Anchor are acceptable. Experimentation is an important part of DeFi and Crypto, and it is one of the reasons it is so exciting. The issue here was how UST was communicated as a stablecoin similar to a fully asset-backed stablecoin like USDC. One of the issues that many people were unaware of was that UST had a completely different risk profile than USDC. The same is true for lending protocols. We must exercise caution and ensure that people understand the risks involved.

Aside from that, DeFi, in general, needs to become simpler to understand and apply. Currently, it’s mostly crypto-savvy people because DeFi is still too complicated and unfriendly for the general public to use. That is why the mainstream primarily uses CeFi such as Celcius, exchanges, and so on, and we discovered that this did not work well. I believe we need to be better at communicating and creating a more friendly and simpler user experience.

Following those events, the market has gone through some healthy development. People have occasionally told me they are interested in Clearpool because it fits the current real yield narrative. It’s amusing when people talk about a real yield narrative because it makes you realize that many protocols’ yields were not real at all and were primarily driven by tokenomics.

When looking at any DeFi protocol, the most important thing to remember is to overlook the token and focus solely on the product. Determine whether it makes sense as a standalone product and generates revenue and value for participants. If it does, you can now consider tokenomics and what the token does. I’m sure there are better and worse tokenomics, but tokenomics should never be the focus of the project. It’s a very positive development in which many good players are building sustainable businesses and rising to the top.

To round out the AMA, I’d like to mention two exciting Clearpool announcements: Clearpool coming to Polygon and LayerZero bridges. Could you please provide us with more information on this topic and the roadmap ahead?

Jakob: We started with Ethereum simply because it was the leading Layer 1 that had been around the longest and proven to be very secure. Given the institutional picture of our protocol, we believe that security should be prioritized above all other considerations at the outset. We were firm believers in the Polygon and its capability to facilitate more efficient transactions and lower fees.

We saw a lot of increased user adoption in just two weeks after the Polygon launch, and the new Polygon pools are already larger than the Ethereum pool. This significant increase in users was expected, but it’s great that it went so smoothly. Polygon’s efficiency makes it an ideal foundation for us to consider launching additional products we have on the horizon. That being said, there are compelling reasons to consider alternative Layer 2 or Ethereum scaling solutions.

LayerZero is another intriguing piece. It is currently used to bridge CPOOL between Ethereum and Polygon. We’ve always wanted Clearpool to be a multi-chain protocol with different ecosystems on different chains. This is something we can accomplish by collaborating with LayerZero, and there will be more exciting news about this topic coming out very soon.

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Clearpool

Clearpool

Clearpool is a decentralized capital markets ecosystem, where institutions can borrow uncollateralized liquidity. Visit us here: https://clearpool.finance/