Investing in Crypto Lending

10 min readApr 13, 2022

This article focuses on outlining my thesis for evaluating platforms in the crypto lending sector — a space that I am particularly excited about in 2022 given various trends that are driving new opportunities for growth.

2021 was a watershed year for crypto. The number of crypto owners tripled to 300 M+. The price of Bitcoin rose by 62 %. The price of Ethereum rose by 386 %. The cumulative market capitalization of cryptocurrencies crossed the $ 1 T mark — and for a brief period, even stayed over $ 3 T. While 2021 was obviously an excellent year for crypto holders, this bull run also had another significant impact: a previously scattered ecosystem of multiple relatively minuscule companies building financial services around crypto began to see scale for the first time. One direct proxy that affirms this is institutional investment data. The CeFi lending sub-sector has cumulatively attracted $ 1.5 B of institutional equity capital, of which $ 1.2 B+ was raised in 2021. Likewise, the DeFi lending sub-sector has seen $ 360 M in institutional funding, $ 170 M of which was raised in 2021 and after (see company-wise compilation of funding data in the annex).

I like to visualize the overall crypto ecosystem in 3 layers (layer 2 being the one of interest here):

Why Crypto Lending is the most promising Layer 2 vertical?

1) Crypto Lending is a relatively recent development: Among the various crypto financial services, wallets and exchanges are virtually a prerequisite to interact with the crypto ecosystem, much like stock exchanges and bank accounts are essential to trade stocks. Consequently, wallets and exchanges have been around ever since crypto came about nearly a decade ago and have matured / evolved alongside with it in terms of capturing TAM and product innovation / variety.

2) Passive yield is the need of the hour: 2021 effectively marked the transition of crypto from being a novelty to an investable asset class. This has propelled the need for passive yield among crypto investors, perfectly setting the stage for crypto lending as the next big opportunity in crypto financial services (after exchanges and wallets).

3) There is scope for increasing the penetration of crypto lending: The crypto lending space is still a relatively small drop in the broader $ 20 T ocean of collateralized lending. An approximate estimate of the total outstanding crypto loans is anywhere between $ 90–100 B as of Q1 2022 (see estimation in the annex). Consequently, at a sector average LTV of 50 %, the total collateral backing crypto loans can be estimated around $ 180–200 B. The cumulative market cap of crypto today is $ 1.9 T — only ~ $ 100 B of which earns passive yield through lending. At a penetration level of ~ 5 %, the crypto lending sector has significant scope for growth even at today’s market cap.

Areas for Innovation in Crypto lending:

I see three gaps in the state of affairs of the crypto lending sector today. Each of these gaps opens up a few opportunities for innovation.

(A) Need for yield exceeds borrowing demand: Although counterintuitive, and unlike in traditional finance, there is greater interest in earning yield (by depositing) than there is for borrowing crypto. This creates an unusual imbalance where platforms end up having a meaningful portion of crypto assets that don’t get borrowed and mostly end up getting either staked on platforms like Yearn or deposited on automated market maker platforms like Curve Finance. This can bring down the effective yield of the deposit pool.

This opens up two opportunity gaps for innovation:

Reallocate un-lent capital (or collateral) to other yield generation sources
Reduce the gap between supply / demand by expanding scope of crypto loans: (i) Expand collateral options (crypto / non-crypto), (ii) Expand borrowing options (crypto)

(B) Low capital efficiency: Crypto loans are largely overcollateralized (barring reversible, smart contract enabled flash loans which account only for a small portion) at low LTV ratios compared to other categories of secured lending. This can be explained by the strong direct correlation of LTV with price and liquidation risks.

This opens up three opportunity gaps for innovation:

Develop mechanisms to reduce liquidation risk (Ex:Liquity’s stability pool)
Mitigate price risk (accepting stable coins as collateral protects downside)
Offer unsecured crypto loans (rely on on & off-chain identity mechanisms)

(C) High APRs for crypto borrowers: Crypto borrowers are charged high interest rates — especially for fixed rate loans — often north of 10 %.

This issue opens up one key opportunity gap for innovation:

Offer interest-reduced / interest-free loans to users

New Theses in Crypto Lending

The crypto lending space already has a handful of large players such as Aave, Compound, and Maker DAO on the DeFi end, and Celsius Network and BlockFi on the CeFi end. Cumulatively these organizations command a significant chunk of the market share for crypto loans. In this space, more so than in other sectors, liquidity begets liquidity (and credibility begets credibility) because investors are already exposed to significant price risk and are thus less inclined to take on additional platform risk without good reason. The largest platforms have more liquidity in terms of total value locked (TVL = deposits + collateral) and are thus able to offer more stable APYs to lenders and are less exposed to deposit withdrawal concerns. Therefore, it becomes all the more critical to offer a differentiated solution in order to achieve scale in this sector.

Based on the seven opportunity gaps identified in the section above, I am optimistic about four emerging theses in this space with the potential to disrupt crypto lending:

1. Rehypothecate borrower collateral to generate yield for lenders: As the problem of high interest rates becomes an obstacle for rapid adoption of crypto lending / borrowing, platforms are looking towards rehypothecation as a solution. A great example is Sturdy Finance which offers interest-free loans to borrowers. The platform stakes borrower collateral into vetted third-party protocols (such as Lido which offer liquid staking) to generate consistent yield. Lenders are repaid from the yield generated by the staked collateral, and the platform too takes a small fee. Thus, while a platform like Sturdy significantly reduces the cost of borrowing — it does so by introducing third-party risk (staking platforms may be exploited / glitched). There are other interest-free lending protocols too; however, Sturdy differentiates itself by not charging any deposit / withdrawal / borrowing fees (tradeoff unit economics for scale).

But this begs the question — if yield is generated as an average of interest-free lending and staking, what’s in it for a lender to deposit funds onto such a platform (like Sturdy) when one can instead directly liquid-stake his / her crypto on a platform like Lido and take home all the yield? This is where rehypothecation plays a role:

B deposits $ 100 worth of USDC on platform
C borrows $ 50 of USDC and puts up $ 100 worth of ETH (50 % LTV)
Total value locked = $ 150 worth of ETH and USDC is staked at 10 % p.a.
After 1-year TVL is worth $ 165, and C returns interest-free loan of $ 50
C receives collateral worth $ 100 leaving $ 115 in TVL
The platform takes a 2 % fee amounting to $ 2, leaving behind $ 13
B makes a 13 % return, versus a potential 10 % by directly staking it!

2. Accelerate collateral liquidation and hence reduce liquidation risk: In addition to volatility, liquidation risk is a major factor in determining LTV or collateral ratio. If a platform is able to easily liquidate (in a shorter time span) a loan when its LTV is breached, then it would require a smaller buffer of collateral. Some platforms are seeking to mitigate liquidation risk by limiting collateral to just one or two highly liquid cryptocurrencies. For example, Ledn, a CeFi lending company, only accepts Bitcoin as collateral. Likewise, Liquity, also an interest-free DeFi lending protocol, only accepts Ethereum as collateral. Liquity in fact also has a unique mechanism in place enabled by a stability pool that makes it feasible for the platform to offer a fixed collateral ratio of 110 % = 90.9 % LTV ratio, which is substantially higher and more capital efficient compared to what most crypto lenders offer.

3. Differentiated approach to risk management (on-chain vs. off-chain & unsecured credit): There are various risk frameworks in the crypto lending space. Maple Finance, for example, relies on off-chain KYC, vetting of borrowers, pricing of loan, and risk management before processing an on-chain transfer of funds. CeFi lenders largely follow a similar process too (except for the off-chain loan pricing feature). Goldfinch adopts an on-chain approach whereby KYC-verified borrowers can only receive a loan once their request has been audited and voted upon by liquidity providers on-chain. TrueFi, on the other hand, employs a hybrid approach combining both on-chain and off-chain verification: borrowers are whitelisted based on off-chain KYC / AML and thorough analysis of a company’s business; borrowing-rate pricing is done on-chain (through voting). Although TrueFi offers uncollateralized loans, it has an exceptional record of no defaults despite originating over $ 1.27 B in crypto loans so far.

4. Extend crypto loans collateralized by non-crypto securities: As outlined earlier, a significant portion of deposits on crypto lending platforms don’t get lent out. One possible solution is to widen the very scope of crypto lending itself, by offering loans backed by non-crypto securities such as bonds / mortgage-backed-securities. Recently in October 2021, Société Générale, the third largest bank in France, made a unique proposal to MakerDAO for a $ 20 M loan backed by residential mortgage bonds worth $ 40 M. The bonds were issued as OFH tokens on the Ethereum Blockchain to be pledged as collateral by SocGen Forge (a subsidiary of Société Générale) in exchange for a loan in USD from MakerDAO. This experiment marked the first interaction between traditional finance and DeFi and has opened up countless possibilities for further innovation along this avenue. MakerDAO and a SocGen R&D team are currently exploring the possibility of scaling this model to Maker’s userbase for wider adoption.

Predictions for Crypto Lending

As the space evolves, I anticipate the emergence of two dominant business models in crypto lending:

Model 1:

Offers collateralized crypto loans
Rehypothecate collateral to generate high yield => borrower pays less APR
Implements faster liquidation mechanism => better capital efficiency
Relatively less KYC / vetting => improves ease of lending
May accept non-crypto securities as collateral => expands scope of lending

Model 2:

Offers unsecured crypto loans
Lower TVL limits ability to generate yield => borrower pays more interest
Strong on & off-chain KYC / identity mechanisms => slower deployment
May accept non-crypto securities as collateral => expands scope of lending

Model 1 makes most sense for retail-focused crypto lending platforms where scale is key in terms of the number of borrowers (which will require faster deployment of loans). On the other hand, Model 2 is more appropriate for corporate / institutional clients who intend to take out larger loans and where a more rigorous KYC / AML vetting process on-chain and off-chain makes sense. I am more optimistic about companies / protocols where the outlined Model 2 is a significant / dominant source of business for four main reasons: (1) lower cost of borrowing, (2) larger ticket sizes of loans, (3) lower likelihood of default due to rigorous KYC, and (4) fewer customers needed to achieve scale. It will be exciting to see how the crypto lending sector evolves over the next several months as more digital asset investors rapidly engage with the ecosystem.

Annex 1: Sizing the Crypto Lending Sector