Hey all, Ben from ScopeLift here. For context, we built the UniStaker contracts for the Uniswap Foundation and would be involved in bringing this proposal to fruition if it is passed. I love the quality of the discussion we’ve had here so far—really thoughtful and great points raised all around. Unsurprisingly, I’m in favor of this proposal, but several responses have made me think a bit more deeply about my opinion. Allow me to expand a bit on a few points:
On Timing & Attack Incentives
Several responses have indicated the timing doesn’t seem right, and implied the DAO has more time to work out a solution before a governance attack becomes a serious issue. I want to highlight a few reasons I think we should be careful about assuming these kinds of attacks are still far off.
Rapid price changes can swing the incentives quickly. Imagine that, as one thought experiment, ETH price quadruples to ~$13K post ETF launch, while ARB stays the same or drops. It’s not implausible to imagine a scenario where the incentives are 4-8x stronger in only a few months. This isn’t a prediction, but just an example of something that’s well within the realm of possibility and could happen fast.
We also shouldn’t assume an attacker has to buy the ARB they will use to execute the attack. ARB in lending markets can be used for an attack while paying only a few days worth of interest. This means acquiring the ARB needed to execute an attack can be much cheaper than the actual price of the token would imply. (Importantly, offering rewards to users who participate in governance staking, by giving holders a more productive way to get utility from their ARB, will also decrease the amount of ARB available cheaply in these lending pools. This is another way in which staking helps improve governance safety).
Finally, as @Frisson alluded to earlier, looking only at the treasury assets underestimates the upside for an attacker. To a certain degree (with some mitigation from the security council, to be sure) the full TVL of the Arbitrum networks come in to play. An attacker can also gain further upside externally by doing things like shorting ARB, longing assets that stand to benefit from ARB being attacked, draining AMM pools with ARB acquired from the treasury, etc… What this means is that the upside for an attacker is likely to be significantly more than the ETH held in the treasury.
In short, while a naive calculation of treasury assets vs. cost of the ARB token might imply a certain cushion of economic security in the short term, the reality is likely worse, and could become much worse much quicker than one might expect.
Impacts on Token Valuation
I think @swmartin made a really interesting point here that’s worth considering:
If we tie the token to insignificant revenue values, we introduce the risk of people valuing ARB relative to its revenue generation and could face compression of multiples.
What’s implied is that the current value of the ARB token enjoys a speculative premium that might erode if investors have “real” numbers to do calculations on. It’s possible this could play out to a certain extent, but I think it actually highlights why moving in this direction is so needed.
Right now the incentives to hold ARB are purely speculative. As such, they are also fickle. A shift in the narrative around Arbitrum—or even just around crypto or L2s in general—could crater the token price, exacerbating the security issues as discussed above. Rather than fearing providing preliminary rewards to ARB holders will erode the speculative premium, we should be concerned that speculative premium is the only thing providing ARB with value today.
In the long run, this is just not sustainable. I think there are enough investors sophisticated enough to understand that traditional multiple calculations will not look great at first. However, demonstrating that ARB is a productive asset, one that will reward holders for their participation in governance stewardship, anchors ARB’s value in something real and tangible. If we shake out unsophisticated holders expecting short term gains, in exchange for sophisticated investors who can recognize the longer term potential, I would consider that a win.
Defining “active participation” and Gameability
I share a lot of the concerns voiced by others about the difficultly of defining “active participation” in a way that isn’t gameable or ineffective. What’s important here is that the system be designed in such a way as to allow this to evolve over time. I don’t expect we’ll get it 100% right on the first swing. That said, the work Karma is doing here seems like a solid starting point.