Liquid Governance

The rise of DeFi and the billions of dollars worth of assets flowing through smart contracts on Ethereum has produced a growing crop of technical auditors, who scour code for any parameter or permission that can be exploited for financial gain. They’ve been busy lately, although many are just yolo’ing into a mainnet launch.

Ask any auditor what opens up the largest attack vector, and inevitably, the answer will be “governance”. Smart contract risks and exploits can be limited to the existing permissions of the contract, but governance – by definition – sets rules for the rest of the system, so co-opting it is akin to a 51% attack on a base layer chain.

Liquid Governance

Maker was the first project to fear a governance attack from rogue token accumulation. In the migration to multi-collateral Dai, it lowered its governance delay to 0, which meant that an attacker with $41m in MKR could coopt governance and drain all the collateral, according to security researcher Micah Zoltu. Maker later increased the Governance Security Module (GSM) to a 24 hour delay, so there would be time to insulate the system from the attack.

The source of the vulnerability to this attack is obvious: MKR is a freely traded token that also contains governance rights. An attacker does not have to infiltrate the system, just accumulate MKR and acquire the governance power it entails.

Liquid governance means that control of the system can be transferred, and inevitably, traded on the open market. Time locks help protect against flashloan attacks and locking up tokens for governance keeps short-term speculators out, but optimizing an ideal governance system is like nailing jello to the wall.


Curve.Finance (CRV) and yEarn.Finance (YFI) are two of the hottest DeFi projects, both of whom launched liquidity mining incentives and on-chain governance systems within the last 6 weeks.

They were both crucial to each other’s success. Curve has $1.1bn in deposits and $734m of those are from yEarn’s Y pool, which offers efficient trading between yEarn’s interest-optimized version of Dai, USDT, USDC and TUSD. Meanwhile, the YFI token was distributed through the same Curve pool.

But the relationship is changing as the protocols move to on-chain governance. Cooper Turley in The Defiant today:

Curve Finance proposed that CRV holders who lock up their tokens in a voting escrow can multiply their rewards by up to 2.5x, starting Aug. 28.

Token holders of yEarn Finance, which holds roughly $2.5M worth of CRV from its early liquidity provider rewards, decided to take advantage of the new incentive, locking the entirety of their CRV treasury for 4 years (the max duration). 

But Curve’s core team locked up their own tokens too, overpowering yEarn’s and every other token holder in the voting escrow. They’re effectively holding ~71% of the DAO’s governance power at the time of writing.

Curve Finance is now encouraging others to lock tokens in order to dilute their super-majority power, but it’s clear that the Curve team intends to remain in control of protocol governance. Governance “moves” like this will increase and while the dust settles, a couple thoughts:

  • You can’t distribute a token while you’re establishing an on-chain governance system. Compound achieved this, but they actually started on-chain governance first. One of its first proposals was to launch the COMP rewards program.

  • YFI could wield tremendous power over DeFi protocols that launch liquidity mining programs.


  • Thread on the game theory around CRV locking mechanism [Spreek]

  • Steem vs Tron: The rebellion against a cryptocurrency empire [Decrypt]

Chart of the Week: Traders prefer low gas

Great chart from the Formal Verification newsletter, which shows that higher gas prices pushed traders from Kyber to Uniswap. Gas fees can be 3-4 times as expensive on Kyber for a simple swap as compared to Uniswap.

Tweet of the Week: DeFi Legal Team awakens

Two separate threads on some of the evolving regulatory issues in the space. The rise of yield farming has led to a proliferation of token launches and potentially more scams that catch the eyes of regulators. The distribution events themselves will be looked into, but there’s also the DeFi products and services themselves that could come under closer scrutiny.

Odds and Ends

  • Aave awarded an Electronic Money license by UK’s FCA Link

  • Synthetix futures primer Link

  • How to conduct your IDO on Mesa Link

  • dYdX partners with Starkware to offer Layer 2 trading Link

  • 1inch announces token and liquidity mining initiative Link

  • Compound proposal to alter COMP distribution mechanics Link

  • Opyn launches options for YFI and WETH Link

Thoughts and Prognostications

That’s it! Feedback appreciated. Just hit reply. Written in Brooklyn, which is as lively as ever.

Dose of DeFi is written by Chris Powers. Opinions expressed are my own. All content is for informational purposes and is not intended as investment advice.

Announcing Govern This

Dose of DeFi subscribers,

I’m excited to announce Govern This, a newsletter about all things governance. Over the last few months, DeFi has shifted from talking about governance to actually governing. Most notably, MakerDAO unveiled a new plan this month to decentralize, while Synthetix made a similar announcement in December, and just this week, Compound launched a governance portal for its new COMP token. Plus, Kyber, 0x, etc.

These are positive and necessary steps for DeFi. The new governance structures are intended to help coordinate across community stakeholders and make better decisions. These dynamics are influenced by the issues covered in Dose of DeFi, but I believe they deserve their own focused analysis.

Govern This aims to educate token holders and make them better voters. Emphasis will be placed on specific governance proposals and relaying community governance discussions on forums and weekly calls.

Governance is a coordination technology that has helped countries and companies build more than the sum of their parts. Blockchains are also a coordination technology, but for computers, not humans. Govern This will track the development of the melding of these two over the coming years.

Like governance, Govern This is a work in progress. I would appreciate any feedback on format, topics covered or any other suggestions to make the newsletter better. Just hit reply.

The first issue of Govern This is below. Please click here to subscribe.

Thanks for reading,


dxDAO aims to power DeFi protocols through decentralized governance

Gnosis launched a long-awaited DEX last week with batched auctions for low-liquidity trade pairs. The front-end, Mesa.Eth.Link is owned and operated by dxDAO, a decentralized collective that hopes to power other DeFi protocols. 

While dYdX does not have any specific governance plans (yet), this tweet from dYdX founder Antonio Juliano is a common approach to governance.

The tweet at the end of 2018 was in response to 0x and its native token, ZRX. The project was popular but the token had no use case outside of governance.

This governance strategy – build now, decentralize later – is widely accepted in the space and is perhaps best exemplified by the A16Z’s Jesse Walden’s post, “Progressive Decentralization: A Playbook for Building Crypto Applications”, which the A16Z-backed Compound has essentially implemented (more in the section below).

dxDAO, on the other hand, maintains that decentralization must come at the beginning or else the core team and investors will have an outsized influence on the project in formal (token voting) or informal ways (dictators for life).


dxDAO was launched in May 2019, spun out of a collaboration between Gnosis and DAOstack over managing the DutchX platform. dxDAO’s key governance design is separating financial rights to the DAO (DXD) from voting power over the DAO (Reputation). It used an Edgeware-style lock drop to distribute reputation to stakeholders in May of last year. Any user could lock up ETH or an accepted ERC-20 for a month and receive Reputation, which are voting rights in dxDAO, even though it is not a token and cannot be transferred.

Over 400 unique Ethereum addresses participated in the distribution scheme. Gnosis went through a pretty extensive process in July 2019 to “step back” from its involvement in the DAO, and since then, the community and dxDAO have aligned behind a mission of “putting the ‘De’ in Decentralized Finance”.

Following on last week’s launch of Mesa.ETH.Link, dxDAO is conducting a fundraiser or (“DAICO”?) to help fund its new slate of DeFi products, including a prediction market platform (Omen) and a privacy-centric DeFi dashboard (Mix).

Project launch is typically when a project is most centralized. Execution is hard and direction and accountability are important. dxDAO’s approach will be an interesting counterexample to the “decentralize later” trend and may provide insight into new governance strategies.

Click here for more information about the dxDAO fundraiser.

Here’s what is on the dxDAO docket this week:

Compound governance goes live, has it found Market-Protocol-Fit?

Since its founding in 2017, Compound has executed with an almost flawless record: no bugs/hacks, a major protocol upgrade and a big name fundraise (twice).

But all of that has been because Compound, the company, has executed well, but can protocol development and the growth of the platform be sustained with community management? We shall see.

Compound’s governance system could not be simpler. Anyone with at least 1% of COMP can submit a proposal of executable code. COMP holders have a 3 day voting period; the proposal passes with a majority of token votes AND a 4% quorum of all COMP tokens.

The 1% minimum for proposal submission is a good anti-Sybil mechanism but it greatly limits participation by small users. There is delegation, so you could imagine a “proposal petition” where you would delegate your COMP to a proposal instead of signing your name.

Compound is clearly taking the “less governance is the best governance” approach. This has worked surprisingly well with Bitcoin and Ethereum, which of course, do not have any formal governance, but those communities clearly have informal governance systems that make decisions.

The biggest governance question for Compound: who is the community?


Other Internet has an intriguing essay on the emergent order from new blockchain tokens and their communities. It is worth a read. It discusses the emergent iteration that blockchains – as a technology and a community – go through to find a niche, both in culture and product.

While it focuses on base-layer blockchains that launch with a token, the essay underscores the most underrated governance element: token distribution. It quotes an insightful tweet from Eric Wall

Before Bitcoin could harden its code and find ‘Digital Gold’ and before Ethereum found ‘DeFi’ and ships ETH2.0, both needed to find a “a strong community of believers” in order to create a “virtuous cycle between headless brands and infrastructural build-out to progressively realize [their] initial promise.”

Communities are connected through a wide spread token distribution, Bitcoin through cypherpunks and online drugs and Ethereum through a global ICO (what Teo Leibowitz called “The Immaculate ICO”).

$COMP distribution

The biggest “news” has been details about $COMP distribution:

  • 24% – Compound Labs shareholders

  • 26% – Founders, current and future team

  • 50% – Compound Protocol users

There are no explicit plans yet, but the widely held assumption is that the COMP distribution will be determined by the interest earned and paid by users on the protocol since its inception. This is a clever way that only incentivizes more use of the protocol and is hard to game because interests accrues over time.

But the question still remains, what will the COMP community look like and what values will it espouse? Can emergent cultures arise out of Silicon Valley too?

Here’s what is on the Compound docket this week:

  • Governance AMA with Compound CEO Robert Leshner – Robert answered a variety of questions on ETH2.0 (staking yield is of great interest), Chainlink (Compound’s oracle system is better), contentious forks (Compound would signal a preference on chain) and how Covid-19 changed his mind about remote work. They also announced…

  • Proposal: Add USDT Support – announced on the AMA, USDT was approved by Compound users in a poll last September but had yet to be included. The proposal to add the largest stablecoin in the world is the first test for the new governance portal. (Very) notably, the proposal does not allow USDT to be used as collateral, as Compound currently does with wBTC. It’s not clear if Compound wants to be in the largest stablecoin market or not and underscores the governance challenges of straddling both worlds.

Maker and wBTC, a test case for the MIP process

While Maker had planned to spend Q2 moving forward with their upgraded governance process, most of its focus has been on restoring the Dai peg.

For more on how the Maker governance process has expanded outside the core community, check out the previous edition of Govern This.

Here’s what is on the Maker docket this week:

Governance and Risk meeting (April 23)

  • Head of Community Rich started off with a new meme for governance: the path from intent to implementation, discussing how forums, polls and other initiatives are designed to capture the intent of the community, and then “empowered people” are tasked with implementing that, foreshadowing upcoming changes.

  • Half of the call was devoted to the addition of WBTC as collateral with representatives from WBTC, Bitgo and CoinList in attendance. CoinList’s WBTC announcement gives WBTC the type of liquidity needed for Maker’s auctions (“can redeem WBTC in less than 2 minutes and burn less than that”). Most of the discussion revolved around the circular loop from BTC->DAI in times of high volatility. While there was some concern that WBTC liquidity was dependent on acceptance as Maker collateral, most on the call seemed to support the addition. The strongest support seemed to come from the Maker Foundation’s market making team, who is reportedly the largest market maker for WBTC. There’s more in the Maker forum thread.

  • State of the peg – Vishesh’s overview (graphs can be seen here) showed that the peg had come down to $1.01x area but most of the discussion was around the debt ceiling. At the time of the call it was 4 million away from its 90m debt ceiling. Vishesh advocated for a more programmatic lifting of the debt ceiling. Update: Dai hit the 90m debt ceiling Friday evening ET. Should help the peg.

Single Collateral Dai shutdown – the process has begun. A poll passed with May 12 as the official SCD shutdown. Just yesterday, an executive just passed yesterday to make the MKR oracle fee-less, which will help with migration. Many in the community think the migration of debt from SCD will do more than enough to restore the peg.

13 MIPs and 2 sub proposals – Core to the new Maker governance process is the “Maker Improvement Proposals (MIPs), which are modeled off of BIPs (for Bitcoin) and EIPs (for Ethereum). The two sub-proposals are to appoint the Smart Contracts Team and assign Charles St. Louis as the MIP editor.

The 13 MIPs are listed below:

–       MIP1 (Maker Governance Paradigms)

–       MIP2 (Launch Period)

–       MIP3 (Governance Cycle)

–       MIP4 (MIP Amendment and Removal Process)

–       MIP5 (Emergency Voting System)

–       MIP6 (Collateral Onboarding Form/Forum Template)

–       MIP7 (Onboarding and Offboarding Domain Teams for Collateral Onboarding)

–       MIP8 (Domain Greenlight)

–       MIP9 (Community Greenlight)

–       MIP10 (Oracle Management)

–       MIP11 (Collateral Onboarding General Risk Model Management)

–       MIP12 (Collateral and Risk Parameter Management)

By and large, the MIPs codify many of the informal Maker governance processes. There is currently a request for comments period (MIP forum) and there will be an informal poll on Monday, April 27 on whether to proceed with the 13 MIPs and 2 sub proposals. If it’s a “Yes”, than an executive for an official ratification vote would start on May 1 and lasts for 4 days. If it passes, the official governance cycle will begin and the rest of the MIPs will likely be approved from May 4 – 6.

Other Governing Things

  • Synthetix trials incentivzation program to encourage ETH shorts to balance debt position Link

  • PieDAO community call on audit and post imBTC actions Link

  • Coinbase Custody double downs on DeFi governance Link

  • Terra considers punishing validators that don’t vote Link

  • 0x governance proposal to decrease epoch length to 7 days Link

  • The case for DeFi Rate governance Link

That’s it! Feedback definitely appreciated. Just hit reply. Written in Brooklyn where it rained all day. No euchre today, but yesterday was epic.

Govern This is written by Chris Powers. Opinions expressed are my own. All content is for informational purposes and is not intended as investment advice.

MakerDAO governance leaves the foundation

On Saturday, crypto fund ParaFi Capital posted a 900-word piece on Maker’s online governance forum entitled [Action Required] State of the Peg. The post argues that Dai is in a “perilous and dislocated state at $1.02-1.03” that demands “emergency action” outside of the newly implemented MIP process.

We discussed why Dai is trading above its peg last week, but the campaign to return it to $1.00 picked up considerable steam over the weekend with several prominent DeFi voices replying in agreement with ParaFi Capital, including Scott Lewis at DeFi Pulse/DEX.AG and Tom Schmidt at Dragonfly Capital, a major MKR holder.

While a 2% premium may seem small, Tom Schmidt explains how that can make a big difference:

“We’re (anecdotally) speaking with a few teams that are considering switching to USDC away from Dai, and we see this switch already happening through on-chain data. dYdX posted its first day 1 with more USDC-WETH volume than DAI-WETH volume last week. Currencies, even fully-backed ones like Dai, still rely on network effects and strong narratives, which can often unwind just as quickly as they grew.

With this in mind, it’s important to think about the problem itself, which is that ~6-10MM more Dai needs to be minted in order to re-peg. So, where is this going to come from?”

The Maker forums are typically populated by the Maker team and core members of the community; most attend Maker’s weekly Governance and Risk meetings on Zoom. Some in the community reacted defensively, diminishing the seriousness of the situation. They argued macro volatility justified the drift from the peg and that rushed solutions could endanger Dai’s long-term sustainability.

Others in the community, like the pseudonymous LongForWisdom and Maker Founder Rune Christenson, were more open to solutions to restore the peg and increase Dai liquidity, namely slashing stability fees and expediting collateral onboarding.

Just this afternoon, Maker’s Head of Risk Cyrus Younessi posted four polls to gauge the community’s feeling on how to proceed:

  1. Are you in favor of adding LINK as a collateral type to MCD?

  2. Are you in favor of adding ETH with a lower liquidation ratio as a collateral type?

  3. Are you in favor of adding additional fiat-backed stablecoins (such as PaxUSD) as a collateral type?

  4. Are you in favor of significantly reducing the stability fee and liquidation ratio for USDC?

LINK looks like it will be fast-tracked for inclusion and # 2 seems like an odd fix, but Rune’s initial reply to ParaFi Capital may be the most prescient:

I think it would make sense to set all SFs to 0 (temporarily) and onboard as many new collateral types that the community has shown interest in, and that the domain teams state can be easily onboarded (so only erc-20’s). Taking such decisive action would hopefully have the effect of overshooting, taking the dai price below 1 USD, and from there the peg can then be stabilized from below, by increasing the DSR and the stability fees.

It’s clear that Rune sees a fully implemented Multi-Collateral Dai system as the key bulwark against the liquidity crunch that Dai finds itself in the long-term and the quickest way to restore Dai in the short-term. But never waste a crisis.

Important: Dai has been trading at ~ $1.015 on dYdX and Coinbase pro for most of the day. If it dips below $1.01, it may assuage fears of a Dai spiral.

Related: Popular currency blogger JP Koning’s latest in Coin Desk argues that Maker should consider negative rates.

How did we get here?

These are tough problems to solve. There are real long-term risks to account for, and there should be push back to short-term fixes for an ecosystem that hopes to be around for decades.

Taking a step back, the MakerDAO ecosystem features 3 pretty different interest groups:

  1. Dai holders

  2. CDP/Vault owners

  3. MKR holders

While there is large overlap in these constituencies right now, they will trifurcate in the future as the system develops. As Scott Lewis pointed out, CDP/Vault holders are much more sophisticated than Dai holders, which may be acquiring Dai to save in a stable currency, rather than opening a CDP to take a levered bet on the price of ETH.

Meanwhile, MKR holders, are the ones with decision-making power. It’s up to them to balance the interests of all three and create a sustainable and secure Dai ecosystem.

There is perhaps a more important constituency that MKR holders should consider: companies/projects building on top of Dai. Two companies come to mind: PoolTogether and Dharma.

Both are building consumer-facing products that leverage Dai but (almost) completely abstract away Dai/Ethereum. With other stablecoin options, it’s a simple choice of what works better and “better” typically means more liquidity – something that benefits all Maker stakeholders.


imToken exploited on Lendfme for $25m, Uniswap for $300k

A lot to still unfold after the largest attack ever in DeFi. There are good technical and Twitter explainers or articles from The Block and Coin Desk. The tldr is that imBTC is an ERC-777 token and there was an exploit unique to ERC-777 on Uniswap and Lendfme (or DForce) that allowed an attacker to reenter when withdrawing and essentially allows you to withdraw more than you supply.

To be clear, Uniswap (or more specifically, ConsenSys Diligence) publicly identified the re-entry attack a year ago during a smart contract audit and Uniswap never offered support for ERC-777 in v1 (even though imBTC had been listed on their official UI). Open Zeppelin even wrote a blog post in July about the ERC-777 reentrancy attack specifically.

DForce/Lendfme, on the other hand, should have been aware of this issue before they listed imBTC and their risk warnings should have gone to 11 after the first imBTC attack. And the same thing for imToken, which runs imBTC.

The incident highlights two key risk considerations going forward:

  • Listing assets is a core governance feature – Choosing not to have any say in what is listed – as Uniswap does – is also a governance choice, but the recent attacks (and bZx) show that the diligence that Maker applies to new collateral may be worth it.

  • Multi-asset pools increase complexity and risk exponentially – there is a 100x difference in the size of these attacks and that’s because all of the assets on the Lendfme platform participate in the same share-risk pool, as opposed to just the two asset-pools on Uniswap.

The incident is leading to a wider discussion on the safety and security of DeFi and increased calls for “composability audits”, but of course, who pays for these? And who watches them?

Liquidity providers bear the most risk in the current set up but there must be some type of diligence-as-a-service for the DeFi community. There may be something brewing. Regardless, the attack will lead to some thorny governance issues:

Click for poll results.

DEX activity heats up

In what was just a monumental news week, two huge announcements in the land of decentralized exchanges:

  • Gnosis launches new DEX with batched auctions – Targeting large trades in low-liquidity token pairs, low-liquidity trades, Gnosis’s new design executes trades in 5 minute increments by conducting a series of auctions across asset pairings and finding the most efficient routes to execute. TokenTerminal has a nice explainer. Gnosis has been around for a while (originally conceived as a prediction market project) and their unique design shows how new projects can fill a specific liquidity gap. Perhaps most interestingly, the front-end of the Gnosis Protocol, is owned and operated by DxDAO, a decentralized collective built on DAOStack that plans to launch other (decentralized) DeFi protocols in the future.

  • dYdX launches BTC-USDC perpetual swap, offering 10x leverage – dYdX has been hard to classify in the current DeFi taxonomy. It’s the 3rd largest lending protocol after Maker and Compound, but also cracks the top 5 DEX’s by trade volume. Their latest launch reaffirms their core focus: margin trading. The Bitcoin derivatives space has seen a lot of growth over the last year and the timing could not be better after BitMex’s problems last month. dYdX is now out in front in the race to build the “Decentralized BitMex” (followed closely by Synthetix). In its pursuit, dYdX took a page from BitMex. BitMex recognized early on that Bitcoin was a good product to speculate on but it was even better collateral for a digital derivatives exchange. dYdX, meanwhile, is using USDC for margin deposits and settlement. USDC will be stable collateral and easy to liquidate. And, of course, this may be the most serious attempt to corner the “DeFi BTC” (alongside tBTC, renBTC, PieDAO, etc) market and is most direct overture at institutional DeFi. Also, talk about a back door way for Coinbase to compete with BitMex and other exchanges. This may be just as good of news for USDC as it is for dYdX.

Chart of the Week: Tokens during their time of day

Cool chart from Flipside Crypto on when tokens are used during a day to highlight time zone differences. Dai is pretty even, but USDT shows a heavy Asian bias.

Tweet of the week: ETH yields > USD yields?

I’m not convinced that this will be a long-term trend but the relationship between ETH yields from staking will have some type of affect on stablecoin yields, which may lead to discrepancies between stablecoin yields on competing PoS networks.

Odds and Ends

  • Black Thursday losses spur $28 million class-action lawsuit against Maker Foundation Link

  • Andreessen Horowitz aims to raise $450m for second cryptocurrency fund Link

  • First look at China’s digital currency DCEP Link

  • Stablecoins pass $9bn, adding $3bn in the last six weeks Link

  • 0xTracker tool shows how 0x API sources its liquidity Link

  • Ethereum addresses for all major DeFi protocols Link

  • Atomic Loans Raises $2.45M for Bitcoin Lending Mainnet Launch Link

  • Update on EIP-1559 Link

Thoughts and Prognostications

That’s it! Feedback appreciated. Just hit reply. Jam packed edition. Be on the lookout for exciting development this week. Written in Brooklyn (again) where I enjoy the mornings in Prospect Parks and dogs without leashes. Jake and Ben won 3-0 yesterday in Euchre.

Dose of DeFi is written by Chris Powers. Opinions expressed are my own. All content is for informational purposes and is not intended as investment advice.

Governance can’t be avoided forever

The DAO has long been Ethereum’s most formidable experience. Crises early in a startup or movement’s history often have an outsized effect on the culture and values as it develops, and the hack and loss of $50m (when ETH was $15) instilled two key lessons for the young Ethereum community:

1.     Security is tantamount with trust in the protocol

2.     Significant changes to the protocol can be implemented when there is overwhelming, informal consensus.

In order to fix the DAO hack and recoup the loss, the Ethereum community decided to hard fork and return the stolen funds. There was a vocal minority that followed the original chain – now referred to as Ethereum Classic – but the overwhelming majority of Ethereum community members supported the new chain, in part because 14% of all ETH at the time was invested in the DAO (by contrast, there is less than 3% of ETH locked in DeFi).

The DAO hack illustrated the perils and promise of smart contracts, which come with a higher level of responsibility and security. Former MyEthereumWallet and Current MyCrypto CEO Taylor Moynahan, a veteran of Ethereum’s early days, carried the security-first banner into DeFi last month with a talk at ETH Denver and a series of public critiques of the bZx’s risk and security standards.

Meanwhile, since the DAO hack, Ethereum’s reliance on overwhelming, informal consensus has worked surprisingly well, and many expected the rickety governance model to be just enough to get it to the ETH 2.0 upgrade.

This loose governance structure, however, can only solve black and white issues, which is why the Ethereum community is in an uproar over ProgPow, because it can’t value the political contribution of users, investors and developers that are building on top of Ethereum, to say nothing of EIP-1559.

Even with the recent contentious debates, there is little support in the Ethereum community for any type of on-chain governance, but that hasn’t stopped DeFi from pushing forward with it to live up to its decentralized goals.

$COMP token and Compound’s simple governance process

Compound aims to create a governance structure that would allow it to be a piece of financial software that is “permanent and upgradeable by the community”. The $COMP governance structure, which was unveiled last Wednesday, will set a precedent for post-ICO token governance and influence Uniswap and dYdX’s decentralization plans.

Not all details about $COMP governance structure have been released but this is what we have so far:

  • $COMP tokens will initially be distributed to Compound shareholders; $COMP is not a fundraising tool.

  • 1% of $COMP is required to submit a governance proposal, which must be executable code

  • There is a 3 day voting period for all $COMP holders, who can also delegate their voting rights

  • If > 50% and a 4% quorum, then the proposal is queued and implemented after 2 days.

The voting period and 2-day timelock delay are lessons learned from Maker governance, and the 1% minimum threshold is a smart anti-spam measure. Compound only envisions standard proposals that could add a new asset, change an asset’s collateral factor or change a market’s interest rate model.

The Compound governance contract has been audited by Open Zeppelin and is currently running on the Ropsten testnet, but there are two clear unknowns that Compound will have to answer in the future:

1.     What value will $COMP have?

2.     How will $COMP be distributed?

$COMP is not meant to be a fundraising device, but there must be a reason for someone to hold it, some benefit that it brings them. Compound argues that businesses that build on top of the protocol – including Compound and its shareholders – will hold $COMP tokens so they can influence the protocol and protect their business interests.

This sounds awfully familiar with 0x’s original token model, which it scraped in favor of a staking system. That’s not to say it won’t work. 0x’s problem wasn’t bad governance; it was the token price. ZRX token holders saw value in an increase in the token price, not from a functioning platform with a separate revenue line that depends on it. Its token model had to change to reflect the interests of its token holders.

Compound governance’s success depends on who holds $COMP and how it is distributed. Early token holders will establish governance norms and need to place the long-term development of the platform over short-term price fluctuations.

Who gets $COMP? The Compound team and shareholders will be early recipients, followed by developers building on top of the platform, at the discretion of the team. And then it will need to distribute tokens to its users; it will likely use earned interest by address, as it did for additional asset voting, to weight the initial distribution.

Compound can draw on its these core constituencies to form an initial governance community, but as soon as $COMP is freely tradable, people will want to speculate on it and get a piece of the 2nd largest DeFi project.

As such, Compound will do whatever in its power to ensure that velocity of the $COMP token remains low. I expect Compound’s governance and decentralization push to move slowly and include lengthy lock-up periods for distributed tokens.

Chart of the Week: DEX volume peaks

February may be the shortest month, but decentralized exchanges (DEX) had their best month of volume ever, with over $372m in volume according to Dune Analytics. The DEX wars twitter bot put the number at $439m but neither of those include dYdX, which reported $160m in volume in February. The volumes are in line with what centralized exchanges saw in February and is not evidence of a shift to DeFi protocols by existing crypto users but does demonstrate their staying power.

Building secure applications in a collaborative environment

Crypto Twitter and Telegram were abuzz this past weekend because an integration (iEarn) with the promising stablecoin liquidity pool swap Curve experienced a trade with massive slippage, leading to iEarn’s founder leaving the space. The above poll from 1kx’s Lasse Clausen is split 50/50, with half claiming that responsibility lies with the developers and the other half putting the onus on the users. While the current debate may be mixed, the next million DeFi users will surely not take responsibility.

Odds and Ends

  • Synthetix uses Gnosis’s dFusion for SNX auction to restore sETH/ETH peg Link

  • 0x updates on staking, liquidity API and off-chain orders Link

  • Making Maker: February 2020 Link

  • Loopring launches first zkRollup-enabled decentralized exchange Link

  • Top 5 DeFi apps by active Ethereum addresses Link

  • Drip aims to bring Coinbase’s Earn to DeFi education Link

Thoughts and Prognostications

That’s it! Feedback appreciated. Just hit reply. Written in Brooklyn. Today’s sun made me forget the weekend’s cold. I’m in Boston next weekend for MIT Bitcoin Expo, let me know if you’re around.

Dose of DeFi is written by Chris Powers. Opinions expressed are my own and do not necessarily reflect the opinions of others. All content is for informational purposes and is not intended as investment advice.