Unchained Ep. 907 Summary

Debate: Should Stablecoin Chains Have an Ethereum L2 or Their Own L1?
Host: Laura Shin
Guests: Justin Bons (Founder and CIO, Cyber Capital), Houndan Lee (Co-founder and CEO, Codex)
Date: September 19, 2025

Episode Overview

This episode features a robust debate about the optimal architecture for stablecoin-focused blockchain projects: should they be launched as new Layer 1s (L1s) or built as specialized Layer 2s (L2s) on Ethereum? With recent developments such as Stripe’s Tempo blockchain and growing competition between Solana and Ethereum, Laura Shin guides Justin Bons and Houndan Lee through key questions about decentralization, incentives, fragmentation, finality, neutrality, and the future of stablecoins and payments onchain.

Key Discussion Points & Insights

1. L1 vs. L2: The Big Tradeoff

[02:15–05:06]

• Host context: Fintechs are converging with crypto—stablecoins are a likely “onramp,” which has led to a wave of “stablecoin chains” (e.g., Ark, Plasma, Stripe’s Tempo).
• Debated question: Should these chains be new L1s or Ethereum L2s?
• Houndan Lee: Warns against “motivated reasoning” for launching new L1s just to capture valuation or a so-called “L1 premium.” Emphasizes building on first principles and delivering user value.
  • Notable quote: "I would caution against any kind of motivated reasoning to back into a set of justifications to justify building an alt-L1 in hopes of some sort of alt-L1 premium..." [03:16]
• Justin Bons: Agrees on first principles, but argues that Stripe/Tempo’s reasons for not building on an L2 were rooted in a desire for true decentralization and permissionless validation. Critiques the current L2 landscape for centralization and unclear decentralization timelines.
  • Notable quote: "If the option is building on centralized L2s, then I'm glad they're building out their own L1—for the sake of decentralization, credible neutrality, for what this whole movement is supposed to be about." [04:32]

2. Decentralization: Reality vs. Roadmaps

[07:03–08:38, 08:08–11:34]

• Justin: Asserts that the “worst” L1s have more robust decentralization and security than most top L2s, where admin keys can still seize user funds or censor.
  • Quote: "If we look at things now, the most basic L1, like the most crappiest L1 number 200 in market cap—you cannot steal user funds. ... Okay, you can do that on the top 20 L2s right now that make up like what, like 99% of the TVL—like this is crazy." [08:08]
• Houndan: Admits some L2s are poorly run, but stresses rollups (at least the better ones) have a “clear path to decentralization”, grounded in “minimizing the diff” between L1 and L2. Cites Optimism’s philosophy as an example; references Ethereum’s resilience and security.
• On rollup security: Houndan argues that the “right to exit” (users can always withdraw assets to L1) is a legitimate check on L2 authority, even at early “Stage One” rollup decentralization.
  • Quote: "Albert Hirschman talked about voice and exit as being sort of the most important checks on authority. Even in the near term, we can have these properties..." [08:50]
• Justin counters: “Right to exit” is decentralization theater if admin keys or security councils can still rapidly update contracts (i.e., to steal funds or freeze withdrawals).
  • Quote: "If the admin key can instantly do an emergency change to smart contract, they can still steal all user funds and there won't be any time for you to do [an emergency withdrawal]." [10:58]

3. Ethereum vs. Solana—Where Do Stablecoins Belong?

[16:04–19:49]

• Houndan: Codex is an L2 because most stablecoin liquidity and commerce are on Ethereum, driven not by BD (business development), but by Ethereum’s credibility and neutrality (“When somebody thinks crypto … you go Ethereum for sure.”) [16:04]
• Justin: Counters that Solana’s metrics (except stablecoins) are now ahead. Argues that 10B usable stablecoins on Solana are more valuable than 100B “locked up” or unsuable/stuck-on L2s on Ethereum, due to scaling and UX limitations on Ethereum.
  • Quote: "10 billion worth of stablecoins that can actually serve as money and be used for commerce in a permissionless way ... is more valuable than 100 billion in an unscalable system … or locked up in L2s that are centralized." [17:58]
• Houndan: For stablecoin users, Ethereum’s network effect and asset depth make it the clear choice—“there’s $160 billion of stablecoins on Ethereum and there’s like $10 billion on Solana. That’s an order of magnitude.” [17:44]

4. Market Dynamics & Incentives

[22:08–23:38, 25:01–26:19]

• Justin: From a first-principles, user-centric view: Ethereum L1 is slow and expensive; L2s have admin keys (centralization risk); alternative L1s like Solana are fast, cheap, permissionless, and truly decentralized—better for users who value those properties.
  • "Of course they'll use an alternative L1. If they actually care about decentralization, of course they will." [22:45]
• Houndan: Questions why, if so, stablecoin adoption is still overwhelmingly on Ethereum, and challenges Justin on “adjectives” and theoretical arguments unsupported by empirical evidence.

5. Decentralization Pathways & Economic Incentives

[50:01–52:39]

• Houndan: L2s have a progressive roadmap for decentralization (voice/exit in stage one; longer upgrade delays and removal of admin key authority in stage two).
  • "In stage two, look like you’ve got like 30 day delays before doing any kind of upgrade. You can’t pause withdrawals. … That’s pretty good. At that point we’re pretty good." [50:46]
• Justin: Skeptical L2 teams will actually decentralize, as it means sacrificing revenue/tight control; incentives favor not decentralizing.
  • "They have to sacrifice all of their revenue in order to truly decentralize. Do you really think they're going to burn the admin key?" [51:48]
• Laura: Connects these critiques to similar ones previously levied against early Solana by Ethereum advocates.

6. Finality and Payments

[54:00–57:08]

• Laura: Asks about the delay from Ethereum L1 finality (~12–13 minutes) affecting L2 payment chains.
• Houndan: Not currently experienced as a major issue for businesses; L2s offer instant confirmation, which suffices for most, especially in payments.
• Justin: For use cases like onchain order books (centralized limit order books), fast finality is crucial—Ethereum L2s will never be competitive there.
  • "That type of finality is completely unworkable as far as I’m concerned ... There’s a massive demand for large, fast central order books… only possible with extremely fast finality." [56:24]

7. Fragmentation & Composability

[57:14–63:48]

• Laura: Raises the concern of user/asset fragmentation as stablecoin-focused L1s and L2s proliferate.
• Houndan: Claims Codex is the only stablecoin chain as an Ethereum L2, so no fragmentation within Ethereum; differences between competitors’ strategies mean they don’t actually clash much in practice. The “stablecoin chain” buzzword will fade.
• Justin: Disagrees—composability (the ability for dapps to interoperate seamlessly) was Ethereum’s key advantage, now severely weakened by L2 fragmentation. E.g., needing to bridge between different L2s with the “same” stablecoin (USDC) is a UX nightmare.
  • "You lose that atomic composability when you move over to this type of L2 ecosystem. And I think that's a terrible trade off to make ... it's a nightmare compared to just scan QR code." [60:08]
• Both agree: Future is in specialized rollups, but Justin thinks only a scaling L1 with seamless composability can be truly “unstoppable”; fragmentation is inevitable otherwise.

8. Specialization vs. Generalization

[66:09–69:46]

• Justin: Ideal world is an L1 which is massively scalable/decentralized and executes all use cases natively—no fragmentation, maximum composability, no distractions from L1 dev teams. Let private parties build L2s if desired, but L1 should focus only on scaling, decentralization, latency.
  • “All the core team needs to do is focus on scalability, decentralization and latency, right? They should have created as much fast, cheap and decentralized block space as they can.” [68:51]
• Houndan: In practice, L1 foundations have to contend with cross-purpose tradeoffs; teams attempting to do everything end up serving none of them optimally—hence trend toward specialization.

Notable Quotes & Memorable Moments

• Justin Bons (on L2 security):
  "If the admin key can instantly do an emergency change to smart contract, they can still steal all user funds … that's kind of just decentralization theater." [10:58]

• Houndan Lee (on L2/Ethereum advantage):
  "Ethereum has had so much adoption because of the enormous credibility and neutrality that Ethereum has. When somebody thinks crypto ... you go Ethereum for sure." [16:04]

• Justin Bons (on user choice):
  "If first principles matter ... you can choose between an unscalable system that’s expensive and doesn’t work, or ... a centralized system ... or ... a decentralized system. When we say first principles matter ... of course they'll use an alternative L1." [22:45]

• Laura Shin (on blockchain evolution):
  "The history of literally every blockchain, including Bitcoin, is that it started centralized. ... This is kind of just how a lot of these blockchains [begin]." [34:22]

Timestamps for Important Segments

• Introduction to debate, context on stablecoins & chains: [01:46–03:16]
• Opening arguments: L1 vs. L2 for stablecoin projects: [03:16–05:06]
• Rollups and decentralization promises/challenges: [05:06–12:20]
• Ethereum vs. Solana for stablecoins: [16:04–19:49]
• "Right to exit" and rollup checks on authority: [08:50]
• Economic/layer incentives—revenue, control, and decentralization: [22:08–23:38]
• Finality concerns for payments: [54:00–57:08]
• Fragmentation and composability problems: [57:14–63:48]
• Specialization vs. generalization in chain design: [66:09–70:27]

Additional Key Topics

Stripe/Tempo’s L1 Choice

• Justin supports building a custom L1 for decentralization and flexibility but would prefer using a proven pre-existing L1 over rolling a new chain.
  • "I actually applaud their decision. ... Personally I would've built on maybe a pre-existing L1." [31:13]
• Both guests discuss the analogy to the early Internet, where private “intranets” lost to the permissionless open Internet—implying open, neutral blockchains will outcompete "corporate" chains.

New Trends: Protocol Native Stablecoins

[74:00–76:53]

• Houndan: The explosion of chain-native/protocol-native stablecoins is the market sorting itself out. Eventually, value will accrue to a dominant substrate, but in the near term, “5 million different stablecoins” may emerge.
• Justin: Competition is good; eventually, liquidity will consolidate. Warns about “decentralization theater,” as in the recent Hyperliquid governance vote.

Takeaways

• Houndan Lee advocates for Ethereum L2 specialization, leveraging Ethereum’s neutrality, deep liquidity, and ability to quickly iterate within a focused market vertical (stablecoins/payments).
• Justin Bons remains skeptical of L2s’ real-world decentralization, highlighting persistent admin-key risks, the superior UX, and composability of properly scaled L1s (e.g., Solana), and the economic realpolitik inhibiting meaningful decentralization.
• Fragmentation is a real and growing concern; both sides acknowledge specialization will win, but differ on the optimal substrate for future innovation and user adoption.
• Finality, neutrality, and credible decentralization remain unresolved challenges for both Ethereum and its L2s, and newcomer L1s targeting stablecoin commerce.

Summary Table: L1 vs. L2 for Stablecoins

| Criteria | Ethereum L2 | New L1 (e.g. Solana, Tempo) | |--------------|-----------------|--------------------------| | Security | Inherits L1, but admin keys persist (for now) | Can be permissionless, but new attack surfaces | | Decentralization | Promised/progressive; not yet reality | Possible from start if well-designed | | Composability| Fragmented across L2s | Native L1 can be unified | | UX | Instant confirmation, but L1 finality lags | Fast finality possible | | Incentives | L2s must cede revenue/control to decentralize | L1s own their economics | | Stablecoin liquidity | Dominant on Ethereum | Growing fast on Solana and others | | Specialization | L2s can focus tightly | L1s can cater to broader markets or do custom verticals |

Conclusion

This episode provided a deep and often fiery look at the tradeoffs between stablecoin-focused L2s (especially on Ethereum) and “roll your own” stablecoin L1s, through the lens of decentralization, incentives, composability, and real-world use cases. Listeners gained insight into both the technical and social/economic factors driving builders’ decisions.

Memorable closing thought:
“If first principles matter ... of course they'll use an alternative L1.”—Justin Bons [22:45]

For further exploration:
See the show notes for links to related projects and debates around stablecoin chains, L1/L2 architectures, and protocol-native stablecoins.