treasure, what's yo pleasure
crypto is a, uh, depending how you dress her
“A man who knows how little he knows is well, a man who knows how much he knows is sick. If, when you see the symptoms, you can tell, your cure is quick. A sound man knows that sickness makes him sick and before he catches it his cure is quick.”
― Lao Tzu
I started sketching out notes for this blog back when 3AC was blowing up, the Terra/LUNA ecosystem was in full-on meltdown mode, and FTX was still a beacon of stability in the crypto space.
In hindsight I wish I had focused more on synthesizing these thoughts then and published this before the events of the past few weeks. Too late now, but as with trees the best time to plant one was 20 years ago…the next best time is now.
If there’s anything the past few months in crypto should have taught us, it’s that we should never assume people are acting within the bounds of a sound risk management framework. This may be cynical but at this point in time it’s the approach we should all be taking. Some of the most “sophisticated” organizations have shown an almost incomprehensible lack of understanding of risk.
Every seasoned crypto person I know maintains the most important lesson is always just survive. You can’t play the game if you have no chips left. Many of the people & institutions that have since blown up preached a similar gospel. The problem is, they weren’t practitioners of their own word.
One of the dirty little secrets is that surviving isn’t nearly as complicated as people make it out to be. The golden rule is never put yourself in position to be a forced seller.
But alas, simple does not mean easy.
So, if allegedly sophisticated investors & institutions can’t seem to get risk management principles right, how do we expect supposedly less sophisticated founders/teams to do so?
It starts with sound treasury management.
While treasury management on its face is a mundane albeit necessary function for early-stage crypto projects, we’ve seen all too well that neglecting this focus can be life-threatening. How many projects died because they allowed 3AC to “manage their treasury”? How many dodged that bullet only to see ½ their treasury evaporate overnight sitting on FTX? The ripple effects are yet to be fully understood but suffice it to say that many of these early-stage projects aren’t positioned to survive an instant 50% haircut to runway1.
One of the unique problems crypto projects face is that most fundraised in their native token. This introduces a very idiosyncratic risk and adds to the reflexivity of the ecosystem broadly. When markets are pumping and capital is cheap & plentiful, nobody pays much attention to the flipside of this coin. But when the tide turns, the native token concentration becomes an existential problem: few people I know2 instinctively denominate in anything other than USD. And as we know, when markets break down, smaller-cap tokens (read: early-stage startup project tokens) move on a levered basis to majors3. So treasuries get slashed twice as hard: crypto majors against USD and then again in the native token on a relative basis against the majors. That can drastically shrink the runway these projects think they have – I’m not necessarily blaming them perse as it’s very easy to anchor to whatever the current reality4 is. All of us can spin narratives to fool ourselves.
So this unique problem is one that needs special attention. One easy solution is for projects to raise in USD and give early investors token warrants. The industry has already moved this direction. There are tradeoffs here obviously and it doesn’t really solve the problem when a token invariably launches. Tokens are supposed to align incentives5 so treasuries default to holding their native token in size. They are signaling to users that “we both want the same number to go up!”.
*It would be disingenuous if I didn’t note there are governance reasons that also influence this dynamic, some of which could be solved through token-class innovation but that’s a blog for another day6*.
This is all well and good theoretically speaking. But the world we live in is far less accommodating and far more complex. The incentive alignment we optimize for should be rooted in long-term value creation (years not months). It’s hard to compound value over time if you get blown out early because you mismanage your treasury.
While it makes sense that *some* portion of a project treasury hold native tokens, the existing landscape is far too concentrated, leaving many unnecessarily overexposed. Long-term alignment means constructing a treasury that can withstand volatile, difficult times – does this mean the treasury “underperforms” during bull markets? Very likely yes. But these projects aren’t hedge funds trying to create alpha through their treasury, so this shouldn’t be their focus7. Take a peek at the top treasury tokens held by each of the 10 largest DAO's (source: DeepDAO) and see if you notice a pattern.
I won’t pretend I know what the optimal native-token-allocation as a percentage of treasury is8 but I’m confident it’s not 70-90%. One of the other pitfalls here is the dual liquidity-faith problem you run into when you do need to sell some of these tokens. For one, no matter how well intentioned you are, the market will view native token treasury sales as concerning. Even if it’s necessary for roadmap development, community tokenholders will *generally* grumble9. (note: If you’re building something meaningful that delivers real value to people then this shouldn’t matter in the long-term anyway). What compounds this friction though is if the treasury is already a massive owner of native tokens, then the liquidity of the market it sells into is extremely thin. The common result here is the project rather bluntly sells down tokens which has an even more severe impact on the native token price. This further inflames the community and has the reflexive impact of reducing the project’s runway since the value of the treasury’s remaining assets are impacted. These are not the feedback loops we like...
Treasury liquidity is important for several reasons: you need to fund ongoing operations, pay contributors, and earn yield so you can match future liabilities. Overconcentration in your native token only amplifies undue liquidity risk. But these elements all tie together; small-cap treasury tokens aren’t ideal for earning sustainable yield and so treasuries that hold an outsized share of native tokens have limited yield-generation options. The other factor here is governance which I’ve mentioned but will leave for another day. I’ve tweeted enough about my gripes with existing governance models.
It’s clear there are complicated factors for treasury management and that these need to be thought through more deeply. Frankly, the lack of sophistication across the board has given me added pause. It’s also made me revisit an idea that a dedicated treasury management provider would be especially valuable in crypto. What could this look like in practice & what problem(s) does it solve?
Ok for illustrative purposes let’s give it a name. Something meme-able clearly, so we’ll call this project TreasuREEEEEE.
TreasuREEEEEE’s value prop is that its sole focus is optimizing your early-stage project’s treasury management function. This is obviously a fluid relationship as startup’s often need to pivot, their development needs can change and timelines are often malleable. Presumably much of the legwork occurs upfront when TreasuREEEEEE is brought on; they’ll be privy to the high-level funding roadmap as well as understand liability needs over the short & medium term. There’s also a discussion to be had regarding risk appetite here but as I’ve made clear earlier, treasuries should not be treated like hedge fund vehicles. There are lots of ways to generate mid single-digit yield and the way you do depends on the composition of the assets you hold. Conservative, long-term focus should be the goal regardless10. With this in mind, working with a provider that exclusively focuses on treasury optimization should free up core team members to focus on building their product/service. An example of different treasury compositions:
TreasuREEEEEE could either provide client custody themselves11 or more likely generate dashboards that abstract away the work TreasuREEEEEE is doing behind the scenes and present treasury data in an easily digestible format to founding teams. These are just possible options but I’m sure there are far more creative ways to execute this in practice.
The argument against this type of service is essentially:
“isn’t managing your treasury akin to managing the capital you raise as a traditional startup…shouldn’t the founding team be capable of that?”
I think yes and no. Without a native token, you can make the case treasury management should be relatively straightforward and similar to the non-crypto world12. But the whole reason I’m writing about crypto treasury is because of the complications native tokens introduce. The ethos of crypto includes trust-minimization and distributing ownership, so it’s important for projects in the space to leverage the existing on-chain stack and build on top of it. But as we’ve increasingly seen, it’s unrealistic to expect those building new crypto primitives to stay up to date and engaged with yield-optimization strategies, vote-farming tactics and other forms of what Michael Dempsey calls “Adjacent Governance”.
Larger DAOs and protocols have moved in the direction of creating dedicated treasury committee’s which at least devote specific resources to this area. It’s a step in the right direction but is still subject to deficiencies around long-term alignment, governance farming (i.e. members elected/removed based on borrowed governance token vote) and a general bias toward inaction. This is also just not a feasible solution for very early-stage teams that are far leaner with little time for ongoing cash management: these teams want to experiment/build/ship, not think about constantly rebalancing their treasury. I’m not suggesting teams that raise money should be ignorant to the high-level importance of this discipline, but there does appear to be an opening for someone to build some type of “on-chain treasury management-as-a-service13".
The ways in which protocols can diversify their treasuries include:
Sell native token for ETH/BTC/USDC/USDT/DAI
Acquire tokens of strategically important adjacent projects
Earn fee revenue denominated in other tokens which accrues directly to the treasury
Yield farm
These types of strategies should be deployed to varying degrees, but it will be critical for protocols to communicate and substantiate decisions. It’s also imperative that a near infinite time horizon is considered when allocating the bulk of the treasury assets. Remember – especially for the earliest stage startups – survival in crypto is half the battle. As someone engaged in the space full-time, I’m acutely aware of how fast it moves so I know how unrealistic it is to expect teams building actual products (something I am not doing) to stay on top of everything going on in DeFi.
*An amusing anecdote*: a couple weeks ago an investor hit me up with a seemingly innocuous question about a defi pool. I looked at it for a bit and couldn’t really give what I thought was a sufficient answer. So I asked a couple people I know who are also regular DeFi users to see if maybe we were missing something. Everyone essentially got back to me with the same answer:
This further cemented my belief that even fewer people than I imagined actually understand what’s going on in DeFi. This is both sad and exciting — it means there are a lot of inefficiencies that still exist and a lot of services yet to be built. Allowing those interested in building new primitives to focus as much of their energy and time on the building part could lead to quicker compound gains for the space. I genuinely believe a team that excels as a pseudo subject-matter expert when it comes to treasury services could unlock a lot of those gains.
As always, dm’s are always open @0xsmac on twitter.
Including a few useful resources teams can use to patch together some type of treasury management solution today:
on top of the declining value of these treasuries thanks to general market conditions
including those who’ve been around for multiple cycles
specifically referring to ETH & BTC here
don’t forget ppl believed BTC couldn’t drop 70-80% this cycle because “too much institutional adoption” exists now
very few projects have what I would consider good tokenomics so take this with a grain of salt, but that is the goal/idea nonetheless
i’m sympathetic to the argument that greater decentralization earlier in a project’s life leads to more vibrant community development and a greater chance at capturing network effects, but I also struggle with this given the current governance token voting model that exists. my default stance at the moment is centralization early with an aggressive plan for distributing the token broadly (if there is a token) makes the most sense
incidentally it’s a bit ironic the straw that broke the FTX camel’s back was being overly expose to its native FTT token
though i have opinions :)
have no fear though, this isn’t unique to crypto. recently, Brian Armstrong had to defend his Coinbase share sales because people freaked out — if you were one of these people i’d recommend checking out Rule 10b5-1)
hippocratic oath comes to mind: first do no harm
fully transparent cold-storage with core project team signatures required for movement above certain $ thresholds…using trust minimized on-chain stack would be best
for the record we already see cash-management-as-a-service type offerings here too (see: vesto). also anectodally i know of big tradfi cash mgmt teams who are being approached by your favorite banking-for-startups to provide similar service
please dm me any & all examples of teams doing this already – I am aware of 1-2 but would love to hear what others ppl find useful/interesting