Why I’m writing this
I’m writing this post because there’s some things I feel like I need to get off my chest and I’m not sure how else to do it. The past few weeks have been extremely stressful in dealing with the decentralized governance process. While I love this entire community, I think it’s time to take a good hard look at ourselves. Something @prose11 wrote last week stuck with me; that the forums and chat are an extension of the workplace and that we must therefore enforce the standards we expect in any workplace. I fully agree with this line of logic and hope everyone else feels the same. But to expand on this point, a workplace is not a social club, it’s not a place that we go to chat and play games - it’s a group of people with a unified goal and the prioritization of achieving that goal comes before all other niceties and benefits. I’ve been involved with this project practically since its inception, first as a community member, then as part of the Foundation, and now as a community member again. I’ve been a MKR holder for almost that long as well. I am sharing this because I’ve been around the block in decentralized communities, both with Maker and other ethereum-related activities, and at the risk of sounding dramatic, if we don’t get our act together quickly things are going to start falling apart. So with that, I’d like to say a few things:
(1) On Core Units
MakerDAO is no longer a startup gasping for air, it is a fast growing application with product-market fit and non-blockchain users that depend on it, and it needs to operate with a level of professionalism that reflects that position. The Core Units we are onboarding are often very talented, but the budgets are seriously frightening to me as an MKR holder. I will not mention any specific names or units, but I find two key disconnects: (1) some Core Units are applying for very high budgets and do not have the commensurate experience, (2) other Core Units have great experience but are effectively negotiating against themselves for their compensation – while neither of these may be an issue in the short term, without some kind of financial oversight and negotiation on behalf of MKR holders, this is going to spin out of control. My personal suggestion would be to create a Finance Core Unit which passes a global budget through MKR holders and then allows individual Core Units to negotiate within that framework.
In addition to currently having a limited grasp on the scope and size (given the lack of clarity around MKR payouts) of various core units, we lack any meaningful way of replacing them or their facilitators. In this kind of environment everything will be drawn to its logical conclusion - “if you don’t like it, try to fire me.” Given the overall friction of the governance process, I can assure you that this is a recipe for disaster. There needs to be a better way to deal with errant Core Units and mandated actors, otherwise we’re operating a political party, not a business. Perhaps it hasn’t happened yet, but one day we will onboard the “wrong” person for a job. If we can’t get rid of them or scrutinize their work in public because it hurts people’s feelings and causes friction in our relationships, we may as well give up before we even begin. My recommendation is to onboard a Core Unit which is tasked with sourcing independent reviews of the other Core Units and making recommendations around their performance and compensation. As an idea, it would be nice to see these independent reviews come from reputable consulting firms that have experience in the area of expertise of the Core Unit and are not otherwise conflicted. These firms could also be tasked with recommending compensation and scope for newly proposed Core Units, or reviewing proposals of this nature.
Finally, we need to prioritize redundancy and decentralization in Core Units themselves. This is a key and missing component in the overall decentralization of the system. If a Core Unit can not effectively be fired because it has siloed information, legal agreements, or IP, then we are going to find ourselves being dictated terms, not negotiating.
(2) On Real World Assets
Since early 2019 I have been openly endorsing entering real world assets into the Maker Protocol. I will reiterate here that it is the only way to scale in the short to medium term unless we’d like to become a USDC wrapper (there remains an open invitation to provide an alternative that’s not negative interest rates, by the way). Over the years, all parties involved with RWA have put in a tremendous amount of time and money into getting us to where we are today. I’m thankful for them and I hope that they’ll keep working towards this admirable goal. Unfortunately, I have to say that I think things have gone horribly awry. Now to be clear, there is nothing wrong with the “concept” of RWA in Maker, just like there is nothing wrong with the “concept” of a smart contract - I am only talking about the implementation. To draw a comparison, both MakerDAO and The DAO share similar names and concepts, but the poor implementation of the latter doomed it to failure (note I am NOT comparing any of the current applicants to The DAO, it’s just an extreme example to highlight a point). It’s my fear that we’re heading down a route of totally avoidable implementation failure. We need to view the legal agreements and structures associated with RWA just as system-critical as we view the smart contracts which power the protocol. So where do I think things have gone wrong?
(A) The current crop of Centrifuge originators (termed DROP in the MIPs) are not structured in a way that provides fairness or recourse to MakerDAO. We have already been taken advantage of once, as our first DROP originator (New Silver) was able to change a key parameter, which is comparable to the collateralization ratio in other Vaults, without any kind of approval from MKR holders (or even the RWF Team). We will likely be taken advantage of again. To quote Charlie Munger - “show me the incentive and I’ll show you the outcome.” Now I actually do not blame New Silver or Centrifuge for this governance failure, they are (allegedly) well within the rights of the documentation they’ve signed. The issue is that these documents (except in generic form and not pertaining to the individual borrower, and apparently with the exception of the Subscription Agreement for Sebastien’s personal investment into New Silver’s DROP token) have not been published to the forums. A comparison I’d draw for this issue is if Protocol Engineering was refusing to show the source code of our smart contracts before they are voted into the system. Well, more accurately it would be as if a bug was anticipated despite not being able to see the full code, and then exploited, and even then no one did anything about it…but I’ll leave this digression for other threads. I have been raising this as a concern, among other issues with these Vaults such as the Asset Originator’s ability to amend material terms and thus the need for an independent director, as a red flag for months but have been routinely rebuffed or dismissed by the RWF Facilitator. The other issue with these Vaults is that the information flow is entirely siloed within the RWF Team and I believe this exceeds the mandate of not just this but any Core Unit. The primary value add of MakerDAO is transparency, if we eliminate this feature for the sake of convenience we will not be in a good position. My recommendation for how to deal with current RWA collateral is to limit debt ceilings to 5M Dai and the cumulative debt ceiling to 25% of the Surplus Buffer until the following issues are fixed with the structures:
- Ensure that the borrower (the “Asset Originator” in Centrifuge terminology) can only change material terms in their agreement with the explicit vote of MKR holders (just like all other Vault types in the system)
- Insert independent directors into all SPVs that can act as general protection for MakerDAO and for protection from bankruptcy
- Get an opinion from a reputable CPA that the Asset Originators will not have to withhold tax on Maker’s stability fees (and if they do, restructure in a legal but more tax efficient way)
- Further distance the ownership of the SPVs from the Asset Originator to increase bankruptcy remoteness
- Ensure an independent agent of some kind can reliably liquidate collateral for any facility where the asset is not a short-term instrument that converts back into cash (e.g. farmland)
Note: I am recommending 5M strictly out of the consideration of being a good partner to the originators, but this number (if anything at all) should be passed by MKR holders
(B) The RWF Team will be more effective by reducing its scope and clearly defining its mandate. Right now, it appears to me that this team is trying to do everything, and understandably things are falling through the cracks. I’d also like to say that, for the most part, I think they are a great group of individuals with the right qualifications. But they are independently trying to (a) source new deals, (b) assist in the structuring of these deals, (c) price risk, and (d) manage the positions. What I’ve just described is generally under the purview of an entire company, not a small distributed team, and I don’t think the answer to our problem is to rebuild JP Morgan as a MakerDAO Core Unit, thus I recommend keeping the team similar in size but reducing scope. I believe it would be better for everyone if we operated RWA the same way that we operate other collateral onboarding and management. The collateral sponsor (in this case the originator) should come to the Protocol independently (or through a third party like Centrifuge or RWA Co.) and be proposed either by themselves or that third party. Prior to a greenlight, elected members of the DAO should not be having negotiations with the collateral sponsor (I’m not saying they do, by the way). After a greenlight, the RWF Team should intercept the application and ensure that the appropriate structural safeguards are in place (like the list I’ve outlined above), and then run the proposal through an objective prioritization framework. The final output being a risk assessment with recommended stability fees and other parameters. Positions should then be managed on a micro-scale by an independent director (or Trustee) who executes a clearly defined agreement, and on a macro-scale by the MKR holders directly. My request to the RWF Team is as follows:
- Limit your scope to enforcing maximum structural protections and proposing Vault parameters
- Ensure you are only needed for oversight and do not represent a critical intermediary in the operation of any Vault
- Do not become a silo of any information critical to the decisions being made by the MKR holders (e.g. deal-specific loan documents)
- Develop an objective prioritization framework that will demonstrate to MKR holders and collateral applications the standards that will get a proposal to the front of the line
- Develop an objective framework for the structural and financial precautions that a borrower must meet in order to access your recommendation for higher debt ceilings, and specify what those debt ceilings will be at each level of increased structural and financial integrity
(3) On philosophy
We need to move forward quickly and intelligently, not move fast and break things. Those two approaches may sound similar, but the absence of thoughtful risk analysis in the latter has never worked in this space. I’m as tempted as anyone to cut corners and take risks, but I just can’t square that line of thought away with the potential impact on users of the protocol. I’d encourage everyone to scrutinize changes to the protocol through a more conservative lens. Recently someone told me “life is full of risks” when dismissing my concerns around a particular collateral application. While this sentiment may be true, it’s a comment that’s more relevant coming from a psychiatrist, not someone assisting in the management of over $5B. I believe our philosophy needs to be one where we do not waive away any risks, and mitigate all that we can. There will be countless risks that we cannot control or do not anticipate, but to ignore the ones that we can and do is simply negligence. This isn’t a damnation of any group or individual, it’s a humble request that we all adopt this philosophy given what is at stake here.
I’m bringing this all up in the hope that it will resonate and things will start to change - if they don’t, the Protocol will enforce this change with MKR dilution. I’d really like to never get there. With that being said, I’m personally exhausted. I’m trying to balance running multiple businesses and a three month old, and I only have some much more time and energy that I can dedicate to trying to constantly keep track of the Maker Forums and Chat. I believe the only way to solve this is with vote delegation (and hopefully experienced delegates), so hopefully that’s coming soon. Anyway, thanks for reading/letting me vent. Still looking forward to changing the world with all of you.