[renBTC] Collateral Onboarding Risk Evaluation

Legal Disclaimer : This communication is provided for information purposes only. The views expressed here are those of the individual(s) quoted or who present said materials and are not the views of Maker Foundation (“Maker”) or its affiliates. This communication has been prepared based upon information, including market prices, data and other information, from sources believed to be reliable, but Maker has not independently verified such information and makes no representations about the enduring accuracy of the information or its appropriateness for a given situation. This content is provided for informational purposes only, and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisers as to those matters. References to any digital assets and the use of finance-related terminology are for illustrative purposes only , and do not constitute any recommendation for any action or an offer to provide investment advisory services. This content is not directed at nor intended for use by the MakerDAO community (“MakerDAO”), and may not under any circumstances be relied upon when making a decision to purchase any other digital asset referenced herein. The digital assets referenced herein currently face an uncertain regulatory landscape in not only the United States but also in many foreign jurisdictions, including but not limited to the UK, European Union, Singapore, Korea, Japan and China. The legal and regulatory risks inherent in referenced digital assets are not the subject of this content . For guidance regarding the possibility of said risks, one should consult with his or her own appropriate legal and/or regulatory counsel. Charts and graphs provided within are for informational purposes solely and should not be relied upon when making any decision. The content speaks only as of the date indicated. Any projections, estimates, forecasts, targets, prospects, and/or opinions expressed in these materials are subject to change without notice and may differ or be contrary to opinions expressed by others.

Proposed Risk Parameters

Risk Premium: 6%
Liquidation Ratio: 175%
Debt Ceiling: 2 million
Auction Lot Size: 50,000
Minimum Bid Increment: 3%
Bid Duration: 6 hours
Max Auction Duration: 6 hours
Liquidation Penalty: 13%
Dust: 100 DAI

Outline

  1. Overview
    1.1 Protocol Summary
    1.2 Tokenomics
    1.3 RenVM security
    1.4 On-chain fundamentals

  2. Metrics and Analysis
    2.1 CEX and DEX volume
    2.2 Holder distribution
    2.3 Tokens on exchanges
    2.4 Redeeming (burning and releasing)
    2.5 Downside risks
    2.6 Volatility
    2.7 Supply growth
    2.8 Defi presence

  3. Risk Parameters
    3.1 Summary of notable risks or red flags
    3.2 Proposed risk parameters

Overview

Protocol Summary

The Ren project’s main goals are to provide a trustless, permissionless, decentralized movement of assets across different blockchains using a distributed platform called the Ren Virtual Machine (RenVM). The RenVM achieves this kind of interoperability with a network of virtual computers called Darknodes, which provide the necessary computing power to identify and process cross-chain crypto assets movement. At the time of writing RenVM supports cross-chain movement of BTC, BCH and ZEC, with BTC as the biggest locked asset in terms of value. Moving assets is supported between Bitcoin, Ethereum and Polkadot chains.

renBTC is an Ethereum ERC20 token redeemable for BTC on the Bitcoin blockchain. One renBTC equals one BTC minus fees and, as an ERC20 token, can be used in various Ethereum wallets and smart contracts.

Tokenomics

While renBTC itself doesn’t have any special tokenomics we do need to be aware it is tightly bound to the RenVM mechanics and to REN token. The circulating supply and even price of renBTC will be influenced by the RenVM fees and bonded REN token (100K of REN bonding - valued at 30,800$ at time of writing - is currently required to run a Darknode).

RenVM is a decentralized custodian that uses bonding to secure the locked BTC and adjusted fees to incentivize Darknodes participation in the network.

The renproject wiki states “Fees are the main incentive for Darknodes to power RenVM. In return for their work, Darknodes are rewarded with fees that are paid by the user. If the user transfers BTC from one chain to another, the Darknodes earn a small adjustable percentage of that transfer. That is, if BTC is moved by users, BTC is earned by Darknodes, and so on. This helps keep the rewards diverse, and the user experience simple (the user does not need to juggle fee tokens).”

The REN bond of a Darknode represents a commitment to good behaviour and can be slashed when the Darknode is responsible for loss of assets. Slashed bonds can be used to restore lost assets. For security reasons RenVM will adjust the fees to try and keep the ratio of bonded value to locked assets value 3 to 1. This means the circulating supply of renBTC will be affected by participation in RenVM and the REN bonded value. Current locked value, bonded value and fees are best monitored on the Ren project mainnet stats page.

Besides the fees on cross-chain movements, RenVM can also adjust a continuous fee (currently at 0%) to incentivize the network and users to adjust locked value compared to bonded value. The continuous fee is implemented through a decreasing exchange rate between the pegged asset (renBTC) and the respective origin asset (BTC). Though this value depreciation compared to BTC would be very slow, it would affect the general price of renBTC, as it would no longer be a 1 to 1 minus fees redeemable for BTC.

Locked value at end of Oct 2020: $326,682,806.92
Bonded value at end of Oct 2020: $42,989,346.00

Fees in cycle/epoch from mid Sep to mid Oct 2020:

  • Mint fee: 0.2% (raised from 0.1% in end Oct 2020)
  • Burn fee: 0.1%
  • Continuous fee: 0%
  • Generated fees value: $336,032

Source of data: Renproject mainnet dashboard

RenVM security

After RenVM transitions to a fully decentralized network and is governed by its community, the security will rely fully on its consensus algorithm and the adjustable economic incentives, mainly through fees. The Ren project wiki pages summarize their security:

“RenVM makes use of a modified version of the Tendermint consensus algorithm to guarantee the consensus mechanism cannot be made unsafe (or halted), unless 1/3rd or more of the Darknodes are adversarial and coordinating (or offline). The same guarantees are made for execution by making use of the novel RZL MPC algorithm.

To incentivise Darknodes to be non-adversarial, RenVM has a series of slashing conditions that will slash the bonds of malicious Darknodes. These slashing conditions, coupled with economic policies to adjust fees earnt, ensure that it is not profitable to attempt an attack against RenVM. Furthermore, an irrational attacker that is happy to make a loss would still need to have access to a large amount of initial capital as well as a way to collude with large numbers of unknown Darknodes in a short period of time.”

During RenVM phases Subzero and Zero (Subzero at time of writing), the Greycore is controlled by the Ren project core team. The Greycore is a special shard of RenVM network that backs all gateway shards and the coordination shard. The Greycore has two core purposes:

  1. Guaranteeing the safety of assets under management by acting as a second signature on all minting and releasing
  2. Guaranteeing the uniform randomness of shard selection

While the Ren team controls the Greycore shard, RenVM does not need to be over-collateralized to remain secure. Over time, once the economics are safe, the Ren team will pass the governance of this shard to the community.

On-chain fundamentals

As described, renBTC supply and to an extent also price depends on the RenVM mechanics and the REN token. REN token is currently intended to be solely used for Darknode bonding but may also transition to a governance token in one form or another in the future. Currently the protocol is governed by the core team to maintain control in its initial phases.

Current REN supply has 999,999,632 tokens with 145,000,000 (14.5%) in Darknode bonding and 110,445,763 (11%) in the Ren project MultiSig. Other notable holders are Binance, Aave lending pool, Huobi, OKEx.

Source of data: Coingecko, Etherscan

With the rapid changing of locked value and fee adjustments we can only very roughly estimate around 9% of generated yield annually ($42,989,346 bonded value, $336,032 one month fees).

Metrics and analysis

CEX and DEX volume

The only centralized exchange with renBTC pairs (BTC,ETH) is Huobi. The liquidity and volume is currently negligible.

As for decentralized exchanges, we’re seeing most of the renBTC supply available in the Curve liquidity pools with the majority of daily trading volume between renBTC and WBTC:

RenBTC-ETH pair trading is limited to the Uniswap platform:

Source of data: Curve.fi, Uniswap

Holder distribution

With the short time that renBTC is available on Ethereum mainnet it’s understandable there’s not many holders. As we’ve seen, most holders decide to lock it into Curve finance.

Current number of holders: 1,364

Source of data: Etherscan

Tokens on exchanges

The vast majority of renBTC resides on Curve.fi decentralized exchange where users are providing liquidity in the pools that swap between different wrapped BTC tokens.

renBTC circulating supply end Oct 2020: 25,996

of which 77% (19,898) reside in Curve.fi liquidity pools.

Curve.fi has only increased renBTC holdings in the past couple of months:

Source of data: Coingecko, Nansen

Redeeming (burning and releasing):

At any point, users can burn an arbitrary amount of renBTC. At the same time, they must specify a Bitcoin address. The Darknodes powering RenVM will eventually observe this burn event on Ethereum, and after 12 confirmations will create a RenVM transaction to release the respective amount of BTC to the specified Bitcoin address.

Best case redeem scenario:

Time for submitted (with included btc address) ethereum tx to be mined + wait for 12 confirmations (3min) + delay by renVM consensus mechanism (to order/batch etc.) to generate signed bitcoin transaction + time for submitted bitcoin transaction to be mined

(expected redeem interval: 10min-60min)

RenBTC recycling via BTC should be significantly faster than recycling WBTC.

reference: https://darknodes.online/transactions/

Downside risks

As renBTC has only been available in the last half year we’ll instead look at the data for BTC to get a comparison against ETH on a larger time frame. As renBTC is redeemable for BTC it has closely followed BTC price.

Daily pullback comparison between ETH and BTC over the past 2 years:

Source of data: CryptoCompare, CoinGecko

Volatility

We will again use BTC past prices for measuring volatility to get enough data points. As the largest and oldest of crypto currencies BTC is one of the most stable (apart from stablecoins).

volatility scaled yearly
90 days 0.507% 23.56%
1 year 0.781% 73.15%
2 years 0.783% 103.62%

Source of data: CryptoCompare

Supply growth

Even though renBTC is unique in it’s trustless, decentralized custodial mechanics, it serves the same purpose as the other wrapped/pegged BTC tokens - to get BTC (or rather BTC representation) on Ethereum. Looking at total supply of BTC on Ethereum, renBTC is in second place (~18%), after WBTC (~75%), while others are lagging behind.

Total BTC on Ethereum supply end of Oct 2020: 147,434 (0.8% of current BTC circulating supply)

BTC on Ethereum in the last year:

Source of data: Dune Analytics

Defi presence

RenBTC defi presence is currently limited to providing liquidity for trading pairs, and is still mostly absent from lending platforms like compound, aave etc. It was recently introduced to the CREAM finance lending platform.

By far the largest usage is providing liquidity on Curve in the renBTC/WBTC and renBTC/WBTC/SBTC pools.

Summary of notable risks or red flags

  • RenVM and the in-house developed RZL sMPC algorithm that protects the locked assets hold the biggest risk. A flaw in the protocol may lead to loss of assets and greatly harm the Ren project reputation. Ren project does seem to put security in top priority and has undergone several audits, however the RenVM network is relatively young as it has only been active for 6 months on the Ethereum mainnet.
  • With increasing demand of BTC on Ethereum, the RenVM may have difficulties getting to and maintaining the 3 to 1 ratio of bonded value versus locked assets value. This would result in increasing fees or in hindering the Ren project to move the Greycore shard control to the community and transition to a fully trustless, decentralized network. This likely wouldn’t affect the price of renBTC but would affect it’s supply on Ethereum.
  • Current liquidity is mostly available in Curve, a minor fraction also in Uniswap. In case of issues in Curve, keepers and users wanting to trade may need to go through a more lengthy process of redeeming BTC on the Bitcoin network and trading on centralized exchanges.
  • If Ren governance enabled the continuous fee, renBTC would slowly decline in value compared to BTC. This may be problematic if the BTC-USD price oracle is used for determining renBTC price and would need to be adjusted.

Proposed Risk Parameters

Risk Premium: 6%

Liquidation Ratio: 175%

Debt Ceiling: 2 million

Auction Lot Size: 50,000

Minimum Bid Increment: 3%

Bid Duration: 6 hours

Max Auction Duration: 6 hours

Liquidation Penalty: 13%

Dust: 100 DAI

The risk parameters are somewhat conservative due to increased risk we’re seeing with the RenVM network. The project is fairly young, considering its age on the Ethereum mainnet, and is currently quite far from reaching the secure ratio of 3 to 1 for bonded versus locked value. Without a more healthy ratio we understand it may be difficult for the Ren project to continue on the desired plan to get to a decentralized network and pass the governance of Greycore to the Ren community. Ren project and its community will use the adjustable RenVM fees to improve the bonded versus locked value ratio which may impose additional problems for the Maker vaults if the current BTCUSD price oracle is used. Debt ceiling is suggested to be increased only if there is a more suitable solution implemented that protects BTC custody.

Lead Researcher: Jernej Mlakar & Andrej Marolt

Sources

15 Likes

Speaking for the Ren team, we think this is a great evaluation!

We’d like to add a few comments/clarificatory statements:

The Greycore will be expanded already during the current phase (Subzero), to include other non-team node operators. And by phase Zero, the Greycore will be under the governance of the community, i.e. the node operators and REN holders, who will be able to vote to add/remove Greycore members (https://github.com/renproject/ren/wiki/Phases#sub-zero).

Binance just recently listed renBTC, which can provide an additional on/off ramp for renBTC (https://www.binance.com/en/support/articles/90f83bbf2b504ffa832f1b1436b52c9b).

If a continuous fee would be implemented, the BTC-USD price oracle should still be able to be used for renBTC by scaling it by the renBTC:BTC rate. But we also want to highlight that the continuous fee is a tool we would likely consider last as a measure to balance the TVL and the bonded collateral, as it makes the UX for both users and developers/integrators worse.

7 Likes

Thank you @MaxRoszko! Most of the data was gathered in end of October so as you’ve noticed there’s been some changes since then. Thanks for the informative and promising comments from Ren team.

2 Likes

This is the part I don’t get. I fail to see how the economics of having a bonded value 3x higher than the value of the renBTC can work at scale.

If we use WBTC as an example, the value (WBTC market cap) is increasing linearly but the minting/burning volume is quite flat. Therefore fees are stable while the value increase. The yield decreases for dark nodes. At scale, revenues from fees will be quite small compared to the renBTC market cap (and the REN bonded value).

So I will expect continuous fee to be used. If you need to lock 3x the value in the network, you can expect fees to be 3x the risk free rate (whatever that is in DeFi, ETH 2.0 yield or stablecoin supply rate on money market) to compensate the capital put to secure the network.

Who would pay 3x the risk free rate to have BTC on Ethereum?

It does not work like that. The REN team is using a 3x of the value of the BTC collateral compared to the value of REN (so you would need to own more REN than BTC if you have any intention to attack the system). 3x is arbitrary - I assume it will drop as the system security increases.

I’m happy to go into more depth here, as the economic security of RenVM is straight-forward in some sense but requires looking ahead a bit, and we haven’t been the best at communicating that clearly.

To start with some base context:

  • RenVM is a growing interoperability platform. Currently almost all fees RenVM generates for the nodes come from BTC to and from Ethereum, but more source chains are being added (FIL, LUNA, DGB, etc.), and more host chains are being added (Binance Smart Chain, Acala, Solana etc.). And when host chain to host chain bridges go live, any arbitrary token can move to and from the supported host chains, including DAI and MKR which can go to and from other host chains, making arbitrage and yield farming across chains super easy.

  • Every renAsset being minted or burned generates fees for the nodes in RenVM. So the more fees being generated, the more capacity RenVM gets to lock up assets, as more fees earned make REN fundamentally worth more, given the limited supply and requirement to bond your REN to run a node.

  • The increased price of for example BTC should not be a factor in the TVL vs. bonded collateral situation in a sufficiently efficient market, as fees are earned directly in the bridged asset, so if BTC goes 10x the BTC fees for the nodes also go 10x and hence make REN worth more, so the bond collateral worth should follow.

With that said, we can look at renBTC specifically, and your example with wBTC which has an increasing supply while their volume isn’t (I haven’t verified that).

There are basically four kinds of behaviors with RenVM that are relevant for the TVL vs. bonded collateral situation:

  1. Assets being minted with an expectation of staying locked for a longer time (increases TVL and increases volume. Example: yield farming)
  2. Asset being burned with an expectation of not minting back in the near term (decreases TVL and increases volume. Example: harvesting yields, exiting to fiat)
  3. Asset being minted and burned in short time interval (no change for TVL and increases volume. Example: trading, arbitrage, on-ramping for other wrapped versions like wbtc.cafe)
  4. Assets moving between host chains (no change for TVL and increases volume)

The business strategy for Ren team is incentivizing and building for behaviors 3 and 4. And Ren governance is for managing behaviors 1 and 2, by increasing the minting fee to disincentivize minting if necessary, and lowering the burning fee (or even giving rebates) to decrease the TVL if necessary.

Of course you also need a good amount of supply and hence high TVL to make the other behaviors possible (to fill liquidity pools for trading), so we did not disincentivize minting during the launch of RenVM. But now recently we’ve increased the fees slightly as the TVL shot up quite fast, and we are still building out the ecosystem and integrating with others which takes time. The results from this fee increase has worked as expected as the TVL has been going down from more burning than minting overall, without a drop in fees earned for the nodes.

And now with all that said, it should be clear why RenVM was launched as centralized in the beginning, with the Greycore in the team’s control, because without building out the ecosystem you won’t get behaviors 3 and 4, which are important to increase volume and hence capacity for higher TVL.

But the plan is to hand over the governance of the Greycore to the community which can vote to remove it eventually, and in the mean time public nodes will be able to participate in signing transactions with the Greycore acting as a second signing group. So the Greycore won’t have the power to do as it wants, letting community nodes keep the Greycore in check, while the Greycore can keep attackers in check by blocking their attempts to steal funds if the attackers got control of >1/3rds of the nodes in a shard.

4 Likes