Amber Token Insights (JPEGD, EW vs ETH)

Recommendation: Equal Weight vs ETH. JPEGd’s low cost of borrow, ability to innovate and deliver, valuations versus uncertainties such as the yet to be launched xJPEGd , emissions and protocol risks.

Mar 22, 2023

Recommendation:

Equal Weight vs ETH. We balance JPEGd’s low cost of borrow, proven ability to innovate and deliver, attractive valuation versus uncertainties such as the yet to be launched xJPEGd, emissions and protocol risk of liquidating collateral in a full NFT cycle.

Liquidity Venue : Sushiswap (ETH Mainnet)

Fireblocks Support : Yes

Prior Audits : Quantstamp, Peckshield, code4rena (audit contest)

English : Amber Research Token Insights - JPEG (English)

Mandarin : Amber Research Token Insights - JPEG (Mandarin)

Description of JPEGd : JPEGd is an NFT lending platform that allows NFT holders to unlock capital by using their NFTs as collateral (over collateralized) to borrow PUSd and pETH (synthetics). The protocol offers holders the ability to borrow against 12 blue chip NFT collections, with up to 70 LTV at full boost. The protocol offers capital efficiency through an auto compounding service of the protocol’s synthetic assets NFTs if liquidated, are auctioned off to obtain its underlying collateral. JPEGd is notably backed by DefiFrog and Tetranode.

Key Investment Points :

JPEGd’s PMF unique mechanics, will make it a key player in NFT lending as NFTfi grows given parallels to traditional art lending

We believe the overall NFT market is set for growth and NFT finance NFTfi will be a lasting feature. The traditional art secured lending market size (2022 USD 32bn as estimated by Deloitte) and the rise of young collectors with an understanding of NFTs underline structural tail winds for the development of an NFT lending market. Associated problems with the traditional art lending markets include lack of access for retail, replicas, opaque pricing methodologies and counterparty risks. These risks could be mitigated by NFTs through on chain verifiability, value assessments determined through oracle based floor pricing and counterparty risks could be mitigated through near instant liquidity (Blur / Sudo) to regain collateral value thereby lowering the overall risk of lending/borrowing.

Technological developments in NFTfi fare well for the future growth of NFTs. Based on the Global Art Market Report in 2021 (with data sourced from NonFungible) NFT trading was ~USD 11.1bn (~17% of traditional global art) and the value of global art (excluding NFTs) was USD 65.1 bn. With NFT trading platforms, instant liquidity venues, platform aggregators NFT futures and NFT lending protocols; we believe that NFTs and its lending market will experience outsized growth, arguably to that of the incumbent art market, minimally.

JPEGd alongside competitors such as NFTfi, BendDAO, and X2Y2 have existing products in the space and arguably, Product Market Fit (“PMF”). Competitive dynamics reflect an oligopolistic market with each platform having an average TVL of ~20k ETH equivalent of loans (X2Y2 ~27k / JPEGd ~21k / NFTfi ~16k / BendDao 25k), ~USD 140mn worth of loans as of 060323. We believe overall market growth of the lending space to 17% of that of the traditional art lending market, a similar ratio to trading, would be a 70x increase in loans outstanding.

JPEGd’s loan mechanics stand out against that of NFTfi X2Y2 and BendDAO. NFTfi + X2Y2 deploys a peer to peer lending model, BendDAO uses a peer-to-pool model and JPEGd uses a peer-to-protocol model. JPEGd uses NFTs as collateral to issue its own overcollateralized stablecoin (PUSd) and ETH synthetic (pETH). This has resulted in the lowest borrow cost but also lower LTV ratios; without boost. JPEGd uses its treasury assets (CVX) to ensure the PUS / 3pool and ETH / pETH pools have sufficient rewards to entice LPs for PUSd and pETH to maintain their pegs. We view the peer to protocol model favourably due to the speed of loan creation and the relatively lower cost of loans. However, we acknowledge the tradeoffs which mainly include, but are not limited, to the pETH /ETH ratio de-pegging if the collateral (the NFTs) behind the pETH cannot be liquidated due to a liquidity crunch.

JPEGd innovations such as NFT trait specific lending limits (can borrow more pETH against an Alien Punk vs a normal CrpytoPunk) and a two tier lending market to meet increasing borrowing demand for pETH distinguishes the JPEG platform from peers.

JPEGd’s eventual (soon™) fee switch which would benefit token holders and passive demand to unlock higher LTV ratios; emissions manageable?

JPEGd revenue drivers includes 0.5% deposit fee for debt drawn, 2% interest on debt, 5% insurance fee on debt drawn that allows a 72 hours buyback of the NFT in case of liquidation, 25% liquidation fee on debt drawn if insurance is exercised and liquidation profits.

Constrained to the size of the available blue chip NFT collections available on JPEGd, with a 5% penetration rate by JPEGd, LTV of 20%, we estimate that JPEGd could produce revenue of USD 4.3 mn annually, ~4% yield vs FDV. The protocol has hinted at initiating the fee switch for xJPEGd (soon™), allowing platform revenues to be streamed to token holders. At 50% of revenue split to tokens, this implies a 2% APR for token holders. Given the reasonably conservative assumptions, we believe that potential fee revenue is north of our estimates given current market conditions. Based off current ETH TVL on JPEGd, the platform’s penetration represents ~1% of the blue chip NFT market, ~20% market share. We predict further sector growth on the back of NFTfi maturing and NFT adoption; and due to lower borrowing costs, business structure and technological innovation, we think JPEGd’s growth will outstrip that of its competitors.

We believe JPEGd looks attractive with considerable growth prospects given its position as a key piece of infrastructure in NFTfi, and the state of its treasury value. Moreover, as of 12^th Mar’23 at USD 0.000655 per token, all in emissions sum up to ~USD 10mn, seems manageable?

JPEGd’s trading price at USD 0.000655, as of 12th Mar’23, is below Treasury value (excluding protocol controlled JPEG) + capital efficient accumulation

JPEGd’s price at USD 0.000655, as of 12th Mar’23, versus its treasury value looks reasonably attractive. As of 12 th Mar’ 23 JPEG’s treasury value sits at USD 58mn. Stripping out JPEGd tokens, treasury amounts to USD 46mn. Out of the USD 58mn, risk assets stand at USD 29mn (50%) while stable assets stand at ~29mn (50%). With FDV of JPEGd at USD 45mn, after stripping out JPEGd from FDV and Treasury, the protocol trades slightly below treasury value. It means one is getting a business that generates USD 4.3 mn in revenues annually for nothing (given 5% liquidation occurrence and 5% protocol penetration rate), with 50% paid out once to token holders once xJPEGd is launched.

$JPEG passive accumulation is possible by being an LP in the ETH-pETH curve pool and staking it in the JPEGd Citadel which boosts rewards with $JPEG tokens. One can deploy ETH Inventory to passively accumulate $JPEG, capital efficiency (mind the A factor though)? (NFA)

Passive accumulation risk includes ETH-pETH depeg, low A coefficient of the eth-pETH pool, JPEGd’s decision not to use their veCRV voting power to incentivize the ETH pETH pool and JPEG DAO’s decision not to incentivize their Citadel with JPEG rewards (unlikely?).

Risk to thesis :

One off inability to liquidate NFTs in an orderly fashion to obtain collateral during a liquidity crunch could lead to protocol losses. As of 17 th Feb 2023, JPEGd’s protocol ‘health’ using floor prices with scenarios of 61% / 71% liquidation levels and the average pETH borrowed per NFT Punks and BAYC will have to drawdown 40% / 30% in ETH terms before liquidations begin, ignoring any market reflexivity.

Chainlink Oracle Exploit? : As far as we understand and after consultation with the JPEGd team, JPEGd uses an oracle with a proprietary TWAP mechanism with the intention of preventing sudden price changes. This is to offer borrowers protection and not expose them to the full volatility of the marginal price. The oracle and marginal price would eventually converge after a 24 to 48 hour period should there be sufficiently low volatility. As a result, the JPEGd oracle pricing is more resilient than the Coinbase Chainlink NFT Oracle as it doesn’t track marginal prices 1:1.

Peak to trough NFT bull bear cycle could wipe out the entire space like how none of the centralized crypto lenders survived.

Delayed launch of fee sharing mechanism for the JPEG token xJPEG Sub optimal treasury management? Issues with CVX CRV?

Regulatory overhang or specific regulatory action against NFTs leading to reduced interest in the space?

Further DD / Qns / Red Flags

Check around clues on xJPEGd launch and if competitors have plans of offering similar products (trait boost system or NFT index)?

Controversial red flag, depending on background of investor class, as the development team and founder are pseudonymous.

Acknowledgements

@arigatopunk / @Tetranode / @OuroborosCap8 / @rafi_0x / @impossiblefinance / @0xLeez / Dune Analytics / Llamacollege / CoinGecko / Coinbase / Chainlink

Contributors : Robin Lim, Joshua Lim, Abhisheak Mall, Steven Shi

Author
Robin Lim is a trader for Amber Group’s Primary Markets Team.

Amber Group assumes no obligation to update or otherwise revise this material, Amber Group does not represent or warrant its accuracy or completeness and is not responsible for losses or damages arising out of errors, omissions or changes or from the use of information presented in this material. Contents in any third-party sources (if any) in this material are completely beyond the control of Amber Group. As such, Amber Group shall not be held responsible for the accuracy, completeness and legality of the contents of such third-party contents. Any reference to third-party contents does not imply an endorsement, representation or warranty by Amber Group. No liability will be accepted for any loss or damage arising from or in reliance upon the contents of this material or these independent third-party contents provided here.

This material is not directed to or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject Amber Group to any registration or licensing requirement within such jurisdiction.

This material does not purport to contain all of the information that an interested party may desire and, in fact, provides only a limited view. Any headings are for the convenience of reference only and shall not be deemed to modify or influence the interpretation of the information contained.

Amber Group不承擔更新或修改本材料的義務，對於本材料所提供資訊的準確性或完整性亦不作任何聲明或保證，並且不對資訊的錯誤、遺漏或變更所造成的損失或損害負責，亦不對使用本材料所提供資訊造成的損失或損害承擔責任。本材料如果涉及第三方來源的內容，其任何內容都不受Amber Group的控制。因此，Amber Group對該等第三方內容的準確性、完整性和合法性不承擔任何責任。任何對第三方內容的引用並不意味著 Amber Group的背書、聲明或保證。因本材料內容或上述獨立第三方內容引起的或其相關的任何損失或損害，我們將不承擔任何責任。

對於本材料的分發、發佈、提供或使用屬於違法或違規行為，或上述行為需要Amber Group事先登記或獲得許可的任何地區、州、國家或其他司法管轄區，本材料不针对或旨在向上述司法管轄區內的公民或居民（任何個人或實體）進行分發或供其使用。

本材料不囊括相關方所希望的所有資訊，僅提供有限觀點。任何標題僅為方便參考而設，不應被視為對所含資訊的修改，亦不應影響所含資訊的解釋。

Amber Labs

Discussion about this post