When Optimism was trading above 5 billion FDV last June, I was outspoken on Twitter about my optimism that the red coin was severely undervalued.
Optimism generates over $40 million in annualized fees and has just announced the Superchain vision, an ecosystem where chains that choose to join will pay Optimism's sorting fees or profits. In other words, I will pay about $5 billion for a chain ecosystem that includes Base and OP mainnet.
As the EIP-4844 upgrade approaches, expected to occur on March 13, 2024, the value of Optimism, a direct beneficiary, has increased significantly, with FDV currently exceeding $15 billion. Therefore, I think now is the time to review the original investment thesis as the major catalysts are in play.
The more I think about the incremental upside Optimism could potentially get further, the more skeptical I become. Don’t get me wrong, I think Optimism, along with the OP Stack and the broader Superchain ecosystem, has become an important infrastructure in the Ethereum ecosystem. $OP tokens may still do well in this cycle, but I still have some important questions about Layer 2 as a whole:
There is a theoretical "glass ceiling" in Layer 2 valuations
To simply describe the relationship between Ethereum L1 and L2, Ethereum L1 ensures the security of activities on L2. Based on this, the overall value of L2 should not exceed that of Ethereum L1 in theory; because Ethereum's consensus mechanism provides authenticity verification for activities that occur on L2. It makes no sense if a lower-cost chain is used to protect activities that occur on a higher-cost chain; otherwise, why would L2 settle on this base layer?
In theory, L2 and even L3 can choose to settle on any blockchain, which ultimately depends on which features these blockchains want to inherit. For a second layer that chooses to settle on Ethereum L1; this blockchain chooses the security provided by Ethereum validators through the consensus mechanism; it also chooses the liquidity that Ethereum has accumulated, and the bridge facilities that are also protected by the Ethereum consensus mechanism.
This assumption should be considered correct unless “settlement layer as a service” becomes more commoditized this cycle with the emergence of the likes of Dymension, or other general purpose Layer 1s can provide the same feature set that Ethereum L1 currently provides as described above.
The counter-argument to this “glass ceiling” issue is that if any Layer 2 can take off in a way that attracts the next millions of users, then it will become a reality. The added value may eventually trickle down to the Ethereum base layer, which will effectively lift the aforementioned “glass ceiling”. My only skepticism about this argument is that:
Given the current trading valuation of Ethereum (330 billion FDV); I think it will be difficult to push Ethereum to certain levels with crypto native currency alone. Ethereum will need a lot of external capital inflows (such as hopefully from an ETH ETF) to go above some of the valuation targets we have set for this cycle.
“Security Need” or “Money Need” is still a relatively new concept in fundamental crypto investor circles; when it comes to infrastructure investing, it requires this school of thought to become the overarching framework
The accrual of value from Layer 2 back to Layer 1 is typically cut by more than an order of magnitude; this problem becomes even more severe after EIP-4844, as the cost of sending data back to Ethereum will actually be reduced by more than 10 times — not to mention that Layer 2 batches multiple transactions, so in order to process 10 times the amount on Ethereum, the fees paid will be more than 10 times.
Layer 2 wars are essentially cannibal wars
According to the above logic; the collective TVL on Layer 2 is always a subset of the entire TVL on Ethereum, because Layer 2 chooses to settle on Ethereum in part because of deep liquidity. When we hold a bullish bias on a single Layer 2 token, we are basically making the following assumptions:
ETH TVL will still double or triple from where it is now, with more optimistic scenarios assuming it will double from current levels.
ETH TVL currently exceeds 40 billion, having peaked at over 100 billion in the previous cycle; and each Layer 2 would require ETH TVL to be three or four times the previous peak in order to have enough TVL and conduct tens of billions of dollars in transactions; providing enough room for growth to make investing interesting.
Layer 2 TVL, as a subset of ETH TVL, will continue to grow;
Considering the major Layer 2s, including Optimism, Arbitrum, Polygon, and newly created Layers such as Manta and Blast; Layer 2 currently accounts for more than 20% of the total TVL. By investing in Layer 2, we assume that this percentage can be at least several times that percentage.
Back in January 2023, when there were only three “rollups” on the market, this percentage was around 10%; fast forward to January 2024, when there are over a dozen general-purpose rollups on the market, but this percentage has only doubled, meaning that the average TVL per rollup has been declining.
· An extension of this - your favorite Layer 2s (like Optimism or Arbitrum) somehow manage to get more TVL than those new and shiny mega-farms (like Blast or even Manta).
Given the two structural reasons above, I am less bullish on Layer 2 as a sector. I think individual Layer 2s may still do well - but this is more due to idiosyncratic reasons rather than general growth across the sector that will eventually generalize to all Layer 2 technologies; two examples I can think of include:
Optimism - $OP still works well as a proxy bet for the entire hyperchain ecosystem, with investors betting that Base will eventually attract the next millions of retail investors because it is close to Coinbase, or Farcaster successfully beats Twitter and becomes the de facto crypto social app;
Polygon - $MATIC or $POL could go parabolic if they form partnerships with the likes of Japan's Astar or Nomura/Brevan Howard in traditional finance; or if the zero-knowledge proof-driven aggregation thesis does well and achieves atomic interoperability between all zkEVMs;
It’s hard for me to imagine a universe where a single Layer 2 beats all competitors and eventually attracts all the top crypto-native partners like gaming and DeFi protocols simply because it’s extremely good at business development. If that’s not the case, how can we be optimistic about and invest in any Layer 2?
Aggressive token redemption plan
Another important factor to keep in mind is the aggressive emission schedule of these new Layer 2s in the next cycle. This is also why I have a bullish bias on older tokens like Optimism and Polygon in this case, as they have already experienced the steepest part of their emission schedules; of course, in hindsight, this is somewhat reflected in their relatively compressed valuations.
The aggressive monthly unlocking of $OP tokens has been an Achilles’ heel for the token price; but as I said, incremental selling pressure will taper off relative to future circulating market cap;
MATIC tokens are almost fully vested, and by migrating to POL tokens, future annual inflation will be only 2%, which is considered reasonable relative to other PoS chains;
On the other hand, some relatively new Layer 2 tokens will finally start unlocking in the coming months. Considering the funding scale of these chains and their previous valuations in seed and private rounds; it is not difficult to imagine that investors will not hesitate to sell on the market.
Only 12.75% of $ARB tokens are currently in circulation; a massive cliff unlock of over 1 billion tokens will occur on March 15, 2024. Subsequently, over 90 million tokens will be unlocked per month through 2027;
From the way they designed the token vesting schedule, it seems like the Starknet team can’t wait to dump $STRK on the market to retail users, after years of building it (for next to nothing), from the way they designed the token vesting schedule – I’m an e-beggar myself;
Printing money to boost business
What’s worse, in addition to the aggressive unlocking schedule, Layer 2 projects have to continuously issue their native tokens to incentivize and reach partnerships. After all, the importance of the underlying technology is self-evident, and business development has become a key differentiator in this competition.
We saw how Polygon funded $MATIC and established impressive partnerships with companies such as Disney, Meta, and Starbucks. But this led to a massive sell-off of its token and explains why $MATIC is trading very cheap relative to other newly launched layer 2 businesses with weaker business development.
At the same time, as large farms like Blast or EigenLayer offer better risk-returns for staked funds in the ecosystem, we are also starting to see early signs of Optimism and Arbitrum issuing tokens to retain users.
Optimism has completed 3 rounds of retroactive public goods funding and issued a total of $40M in OP (equivalent to >150M) tokens to projects that build and utilize the OP stack within the ecosystem. Arbitrum has also conducted multiple rounds of short-term incentive programs and issued over $71M in ARB tokens to projects; and is even considering establishing a $200M gaming-focused ecosystem fund and long-term incentive program to continue to drive user activity.
It’s reasonable to assume that this aggressive incentive will only continue in this cycle until a clear winner emerges in the Layer 2 race, and until then I think Layer 2 as a category will generally lag behind in terms of price performance.
