There’s a lot of negativity in the broader crypto space best caputred by Dougie Deluca’s “Crypto is dead” post that argues that crypto culture will not be adopted even if blockchains are. That feels right because the decade’s excesses — memecoins, VC L1s, Web3 games, NFTs — have poisoned the well.

But DeFi still feels like the cleanest part of crypto: transparent markets, composable primitives, and a culture that takes the plumbing seriously. If crypto culture fades, maybe DeFi’s memes get louder because they’re tethered to utility instead of vibes.

We’ll see. The world keeps barreling to a financial system run on smart contracts on truly digital rails, but what will it look like? One can only meme.

Solana bros are shilling Internet Capital Markets hard; TradFi has embraced RWAs; Stripe is wielding stablecoins as a trojan horse; or perhaps Coinbase and Robinhood hope tokens just eat everything.

DeFi is still where I’m planting my flag. It’s just cooler.

Anyway, here are the top memes and charts for DeFi in 2025

1. STABLE GENIUS

The biggest victory for DeFi in 2025 was regulatory momentum for stablecoins. The GENIUS Act, signed into law in July, created the first federal framework for payment stablecoins by requiring 1:1 reserves and regulated issuance. It might end up as the most significant bipartisan legislation passed in the 2020s (or the CHIPs Act). That says more about the rancor and stasis in Washington these days but it says even more about the undeniable utility of onchain money. Stablecoins are simple, easy to understand, and carry no speculative baggage.

With the rules set, the adoption phase has finally arrived. We are seeing three distinct categories emerge:

1. Savings & Investment (The Bedrock)

2. Payments (The Growth)

3. Agentic AI (The Frontier)

Issuers love to talk about transaction volume, but the reality is that the vast majority of the $300b now onchain is there to store wealth. Total stablecoin supply remains the ultimate scoreboard for comparing chains. But going forward, the growth story shifts.

The payments layer is poised to explode in 2026, led by stablecoin cards and the “fintech-ification” of crypto. Meanwhile, AI agents using stablecoins makes intuitive sense but still seems a ways off.

What stablecoins have proven is a sustainable business model. Stablecoins turn distribution into deposits: win the wallet and you capture the spread on reserves plus fees on flows. Tether is widely known as one of the most profitable companies in the world, while Circle soothed the haters this year with a public markets valuation 5x as big as the industry valued it a year ago.

Now stablecoins look less like a crypto feature and more like the revenue strategy for a new generation of consumer financial products.

Read more: The Stablecoin Pact: Crypto’s New Order after the GENIUS Act

2. Ethereum L1

For years, Ethereum supporters and haters alike preached of a multichain future, but the story of 2025 was one of consolidation: DeFi apps relearned that being everywhere has a cost, and liquidity increasingly concentrated on a smaller set of chains. Ethereum’s L1 is still standing strong after several years of well-financed (and incentivized) competitors trying to entice liquidity away.

In 2024, “parasitic L2s” was on our list of top memes, but a reboot at the Ethereum Foundation has recommitted the ecosystem to scaling the L1. It shows: gas costs are at multi-year lows even though transactions are at all-time highs. The refocus on the L1 has not greatly affected Ethereum’s L2 ecosystem, either. Base and Arbitrum are major ecosystems in and of themselves, while launching an L2 on Ethereum seems to be the clear winner of the “appchain” thesis.

It seems that Ethereum might be able to have its cake and eat it too. The next few years will see an onslaught of TradFi and Big Tech companies launching a “stablecoin + chain“ combo. There will be more Robinhoods and Celos that see the benefits of being an Ethereum L2 than Tempos and Arcs that go through the trouble of launching an L1.

This consolidation is quantifiable. Ethereum L1 retains a dominant 57% share of stablecoin supply, but the real story is credit: it has $25bn of active loans, roughly 10x its nearest competitors (Plasma and Solana).

That’s not to say Ethereum L1 will be the only game in town (especially as it relates to this year’s meme #4). The pie will be big enough for alternatives but this was a year where Ethereum’s L1 reasserted itself.

3. Risk curators

Risk curators — or, as I prefer, DeFi vault managers — were one of the clearest signs in 2025 that DeFi is maturing. Their lineage runs back to 2020 and Yearn, but the category leveled up once ERC-4626 standardized vault design and modular lending platforms like Morpho and Euler made it easy to package strategies into one-click products. In practice, vault managers have become the user-facing layer of DeFi: they choose collateral, set risk parameters, and route capital across markets so users don’t have to.

2025 was a breakout year for the sector, with total AUM surging from $1.6b in January to a peak of over $10b by late October. However, the market proved fragile. In November, the collapse of Stream Finance triggered a contagion event that wiped out hundreds of millions in value, punishing managers who flew too close to the sun with opaque, yield-chasing looping strategies.

MEV Capital and Re7 took the hardest hits with AUM decreasing 80%+ for both over the last three months. Steakhouse and Gauntlet have emerged from the wreckage as the clear winners with their marketshare sitting at 26% and 21% respectively.

With the hierarchy of trust now established, the focus for 2026 shifts to distribution and the race to capture mainstream capital. Vault managers need to win both institutions (clear mandates, reporting, liquidity discipline) and retail (simple UX without hiding where yield comes from). The most scalable path may look like a “DeFi mullet”: centralized rails in the front and onchain vaults on the back end — plus new channels like L2 treasuries putting bridged assets to work and, maybe even from Digital Asset Treasuries (DATs).

Read more: DeFi Curators in 2025: Navigating Chaos, Building Resilience [Chorus One]

4. Perps

Perps have long been crypto’s most successful product, but they’ve been dominated by CEXs. While protocols like dYdX, Synthetix and GMX innovated with novel liquidity models, they historically struggled to match the high-frequency experience of centralized venues. That gap finally closed in 2025 as specialized execution environments made onchain perps fast enough to compete on feel, not just ideology.

At the beginning of the year, Binance’s daily perp volume was roughly 5-6x that of all onchain DEXs combined. Today, onchain venues regularly flip them.

Hyperliquid kicked the door down by building a high-performant perps protocol, but they’ve been joined by Lighter, Aster and others that have been able to replicate their success.

An astute reader may notice I’m calling these protocols “onchain” rather than “DeFi.” In my book, DeFi requires verifiability, and much of the leading perps stack still falls short: Hyperliquid is not open-source, and neither is Aster. Lighter just made their code source-available, which is a meaningful step toward making performance and transparency coexist.

The category still has sharp edges, and the auto-deleveraging incident of October 10 made that plain. Gauntlet’s Tarun Chitra published research claiming Hyperliquid’s queue-based engine inefficiently destroyed over $650 million in trader profits, arguing for a pro-rata model instead. Paradigm’s Dan Robinson and the Hyperliquid team pushed back, saying Tarun’s model incorrectly assumed total equity seizure rather than standard position closure.

This controversy illustrated the precise danger of closed-source execution: without visible code, the market was forced to debate theoretical models rather than simply verifying the facts.

Ultimately, while perps may be used by gamblers now, the success of these protocols brings the vision of onchain price discovery closer to reality. The question in 2026 is whether equity perps will emerge and whether they find a market outside of crypto gamblers.

Read more: Autodeleveraging, Hyperliquid, and the $653m Debate [Nagu/Dare to Know]

5. DAOs are dead

As a longtime DAO contributor, this has been a tough meme to digest. But the shift is real. DeFi governance is moving away from “DAO-first” ideals and toward control fights and corporate-style consolidation. The latest example is the civil war inside Aave over who controls Aave’s IP and the fees on Aave.com. It is also showing up in the numbers: DAO grant distribution fell to $44m in the first half of 2025, down from $848m in all of 2024.

The same pattern is playing out elsewhere. a16z called for the end of the foundation era this year and Uniswap is moving in that direction with the UNIfication proposal, which just passed a Snapshot vote last week. The proposal folds the foundation into Labs while rerouting economic value from the entire Uniswap Ecosystem into the UNI token.

Here’s the uncomfortable truth: the DAO meme was often used as a token launch strategy. It gave projects a credible claim to decentralization, especially when regulatory pressure was high. As the regulatory winds shift, there is less appetite for decentralization theater.

That is not a bad thing. For years, too much of crypto treated “decentralization” as a loophole for issuing speculative assets. The real point of decentralization is simpler: it changes the market structure. When anyone can build, fork, or route around the network, it creates credible alternatives and forces competition, which is what drives innovation over time.

Before blockchains, we could only introduce this kind of competition at the economy level, using markets and governments to check monopolies. But in the digital realm, those checks hit a ceiling because the infrastructure itself stayed proprietary. Governments can regulate firms, but they can’t make a privately hosted execution environment neutral or verifiable. When a single company controls the servers, the front end, and the rails, they control the rules. Blockchains changed that by enabling execution on a shared, permissionless base layer, extending free-market dynamics into digital infrastructure itself. They are a necessary step toward credible decentralization, but they are not sufficient.

We need coordination structures that align incentives in a network. This was the dream of DAOs but they have failed.

What’s missing is an economic model that lets open-source software capture value, and a network structure that can interface with the real world.

At Powerhouse, we call the framework open-source capitalism: making open-source investable by tying it to revenue-generating hubs and turning profits into durable funding for builders. We call the structure a Scalable Network Organization (SNO): a DAO as the decision layer, connected to operational, commercial, investment, and IP entities so the network can act like a company without becoming one.

DAOs are dead. Long live DAOs.

That’s it! Feedback appreciated. Just hit reply. Just finished before the year is done! Written primarily in Austin, Texas. I hope everyone has a happy New Year.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO/Sky. Some of my compensation comes from SKY, so I’m financially incentivized for its success. All content is for informational purposes and is not intended as investment advice.

It's ironic that the first-ever crypto legislation enshrines fiat currencies into the fabric of blockchains. This is a far cry from the early crypto days where fiat was the enemy, and the separation of money and state was the goal. Fast forward, and now US politicians are talking about how blockchains could further the dominance of the USD.

How did we get here? And where are we headed next? With stablecoins now set to be enshrined in law, crypto's journey from aspiring anarchist money is pointing squarely at its true destination: stateless financial infrastructure.

Blockchain, not Bitcoin

In the early days of Bitcoin and crypto, there was a common sentiment among the suits of Wall Street: they were interested in the underlying technology of Bitcoin (blockchains) more than in Bitcoin the currency. Some even started what they called “enterprise blockchains”, while others dropped the blockchain moniker altogether and said they were building “distributed ledger technology”.

None of those efforts succeeded at attracting usage and were generally scoffed at by those in the Crypto industry (and not just Bitcoiners), because they weren’t permissionless or decentralized. The suits of Wall Street believed that a volatile currency was not going to become the foundation of a new financial system, and eventually us crypto bros figured it out. As we highlighted in (my favorite) Dose of DeFi in February 2020:

There’s a common trope where a Bitcoiner – typically an American male – admonishes an inflation-ravaged country and preaches the gospel of Bitcoin and its fixed supply. From his comfortable perch on Twitter, he speaks of how Bitcoin is the solution to a developing country’s woes, if only the people would appreciate hard money and Austrian economics.

So far, Bitcoin has not been a savior to any economy. Its censorship resistant P2P payments have given individuals in repressed regimes an important lifeline, but it has not been adopted at scale in Venezuela, Zimbabwe or other inflation-challenged countries.

We thought they wanted Bitcoin, but maybe it was just the US dollar?

In the end, ideological skirmishes gave way to market demand. The quiet utility of dollar-backed tokens on blockchains proved to be the pragmatic path for adoption.

Crypto grows up

So, it seems crypto's adolescent dreams of overthrowing the monetary order have largely faded (for now), replaced by the more sober reality of its fundamental role as financial infrastructure. Early Bitcoiners were hyper focused on making the money angle work, but in the end, there were no takers (sin El Salvador). People, as it turns out, were less interested in a new currency and more interested in using these novel blockchain rails to move good old US dollars around the globe, efficiently and quickly.

This proved particularly helpful in the often-clunky world of crypto exchanges as Bitcoin professionalized after the 2013 bubble. Initially, Bitcoin itself served this purpose: you could withdraw it from one exchange and deposit it into another, sidestepping the ponderous bank wire system as long as both platforms supported BTC. Then came Tether (USDT), offering an upgrade by removing the pesky currency risk inherent in using Bitcoin as a bridge asset. Why gamble on Bitcoin's price swings when you just wanted to move value?

A few years later, after the 2017 ICO bubble created assets onchain and thousands of ERC-20 tokens, new DeFi trading and lending protocols launched, adding another layer of validation for blockchains (as genuinely-useful financial plumbing). It soon became apparent that crypto’s most popular product – leveraging-up to buy more crypto – only works if you can borrow in a stable currency. These new DeFi lending platforms provided the first onchain yield for stablecoins, and in doing so, broadened their appeal beyond traders to anyone simply looking to save in dollars, especially if they were outside the traditional US financial system.

The stablecoin ah-ha moment

It’s important at this point to remember that this wasn't some grand, orchestrated plan by the US government. No one told Bitfinex to use a USD stablecoin to solve their banking problems. Justin Sun wasn't acting on a State Department memo when he used dollar stablecoins for his gambling apps in emerging markets. And no one from the Treasury told Uniswap to upgrade its v2, ​​which newly enabled any asset to be paired with stablecoins, not just ETH. This was raw, unadulterated market demand. Blockchains proved to be remarkably efficient financial rails, made infinitely more useful by allowing the world's reserve currency, dollars, to flow through them.

This organic, market-driven gravitation towards US dollar stablecoins is precisely what jolted lawmakers into action. There's always been chatter about crypto regulation, oscillating between suffocating overreach and a dangerously light touch. But the stablecoin bill managed to cross the finish line because the usual crypto advocates found unlikely allies, like Senator Mark Warner (D-VA):

The stablecoin market has reached nearly $250 billion and the U.S. can’t afford to keep standing on the sidelines. We need clear rules of the road to protect consumers, defend national security, and support responsible innovation. The GENIUS Act is a meaningful step forward.

So, to recap: crypto’s big experiment as a new form of money mostly belly-flopped. But as financial infrastructure? It has found its footing, ironically by hitching its wagon to the very US dollar it once sought to supplant. The explosive growth of stablecoins forced the legislative hand, but this bill isn't designed to kill stablecoins. It's the opposite. It’s a legitimization, a welcoming committee ushering stablecoins from the shadowy corners of Degenistan into the brightly lit, compliance-polished plazas of Wall Street and the meticulously-curated walled gardens of Silicon Valley.

Stablecoins in the US: New financial plumbing

Now back to the GENIUS Act, which isn't just about regulating some niche crypto asset. It's laying the groundwork for a rewiring of US financial plumbing, with stablecoins at the center. While there are short-term competitive dynamics for existing stablecoins, the long-term implications are broad, touching everything from how we pay for things, to the very structure of banking.

Payments and the Stripe test: The most immediate impact will be on payments, and we're already seeing major players dip their toes in. Stripe, the leader in online payment processing, has rolled out stablecoin-powered financial accounts – initially supporting USDC – targeting businesses globally. It also acquired Bridge, a stablecoin infrastructure company, which enables some smooth DeFi onboarding experiences. Visa and Mastercard are also in the game, enabling consumers to spend stablecoin balances through their existing card networks, by partnering with issuers and wallet providers.

Bank disintermediation and the rise of ‘narrow banking’: This is where things get really interesting (and potentially uncomfortable for traditional banks). If stablecoins become a preferred way to hold and move dollars, especially for businesses and digitally-savvy consumers, what happens to bank deposits? The traditional banking model relies on deposits to fund lending. If a significant chunk of those deposits moves into stablecoins, held in digital wallets or on platforms like Coinbase, banks could find themselves disintermediated.

The truth is, banks are kind of already disintermediated when it comes to lending. Yes, yes, they do a lot of small business and mortgage lending, but specialized mortgage lenders often originate these loans, which are then funded by larger banks from their balance sheet. In a new world, these mortgage lenders would partner with investment funds that are specialized in evaluating mortgages (or any loan). And so, we unbundle into a loan originator and then an investor. If you squint, this looks like what the large debt markets are now, both public bonds that are freely traded and private credit, which is so hot right now. This shift toward what economists call "narrow banking" – where financial institutions take deposits but invest them primarily in safe assets like treasuries rather than making loans – is being accelerated by stablecoins.

Yet this future may be a ways off, because the banking lobby was able to insert a provision in the GENIUS Act that prohibits interest-bearing stablecoins. While the aim is likely to mitigate risk, the demand for yield is a powerful force. We can anticipate that interest will still ‘leak out’, potentially to less regulated, non-US based platforms and products, similar to how Aave rates can spike due to strategies like looping Ethena’s USDe, or the emergence of offerings like Sky Savings Rate. If these external platforms become significant, this could eventually create new systemic risks for the US financial system, forcing regulators to play catch-up with stablecoins once again.

Huge new opportunities in DeFi (and the question of your next mortgage): So, if banks aren't the primary source of credit, where will you get your mortgage? Or your small business loan? Enter DeFi. Our hope is that DeFi can step in to make this evolving credit system more transparent and efficient. We don't have to use banks to extend credit; we do so now within a heavily regulated market structure designed to ensure stability. DeFi protocols already offer avenues for yield through lending and liquidity provision. Imagine a future where your mortgage isn't funded by a bank but by a DeFi protocol, funded by a global pool of stablecoin liquidity. Yield-bearing stablecoins like BlackRock's BUIDL fund or Ondo Finance's OUSG offer a glimpse. These products provide compliant access to yields from traditional assets like US Treasuries, often for institutional and accredited investors. Could similar structures be adapted for broader consumer lending? This could create a vibrant, though potentially riskier, landscape for credit. A new order with less credit intermediated through banks might be better, but we need to tread very carefully.

Stablecoins outside the US: The dollar’s new dominance

While of course the immediate implications of the bill lie within US borders, it also plants a seed for a shift in the role of the USD world-over. The other irony of the rise of stablecoins is that Tether, (by far) the largest US dollar stablecoin, is domiciled outside the US. While Tether complies with all US law enforcement requests regarding KYC/AML laws, its founders have been skittish about entering the US.

And while American politicians are focused on what happens to American citizens, the GENIUS Act will have an even greater effect on the rest of the world. Whereas for the US, stablecoins’ legitimization will lead to a change in market structure for credit dissemination (that’s a big move!), for the rest of the world, it will challenge the monopoly of local currencies. This isn’t just about finance; it’s about monetary sovereignty.

Fiat currencies: The new local newspapers?

The dynamics here about local monopolies collapsing remind us of Ben Thompson’s Aggregation Theory for media. It’s playing out again, but this time with money. When distribution costs for news collapsed thanks to the internet, most city papers withered and died, while giants like The New York Times went global. The internet rewarded the biggest, most trusted brand.

The same dynamic is about to hit fiat currencies. Citizens, especially in countries with unstable economies or restrictive financial systems, will increasingly prefer USD-denominated stablecoins because they're perceived as more stable and offer access to global commerce. Governments, naturally, will be inclined to restrict them to protect their own monetary control, but can they really? Network effects are powerful.

The Trump factor

This runaway success story for US dollar stablecoins hinges on a critical assumption: that the US wants to remain the global reserve currency, and that increased dollarization is actually good for the American people. The Trump administration, interestingly, seems to think otherwise, actively trying to de-dollarize the world with the so-called Mar-a-Lago Accords. It’s not entirely clear what their endgame is, or how Americans truly feel about this.

But just look at the recent jump in US Treasury yields as the Trump administration's rhetoric spooked foreign investors from holding treasuries. Lower global demand for dollars and treasuries doesn't just mean more fiscal headaches for the US government; it translates to higher mortgages and car loans for everyday Americans. So, while the crypto world is busy building on-ramps for global USD stablecoin adoption, the question remains: is this truly the desired outcome for the US itself? The GENIUS Act might officially be about regulating stablecoins, but it's inadvertently supercharging a much larger domestic conversation about the dollar's future role.

Stateless money vs. stateless financial infrastructure

While crypto is moving onto stablecoins and stateless financial infrastructure, it will never fully abandon its dream of monetary sovereignty.

Bitcoin has definitely achieved stateless money. It marked another all-time high this week and still has a long way to run as digital gold (even if it’s not toppling dictatorships). Other forms of stateless money will likely emerge. Some would argue – and we’d agree – that Ethereum has reached that level precisely because it is such a core stateless financial infrastructure for so many stablecoins and RWAs. Solana may get there as well, but all crypto assets that achieve stateless money status will do so both through fee generation – or REV as the cool kids call it – and a monetary premium that is the collective belief in the future of the network by tokenholders - hey, not so different from Jay Powell's regular pep talks for the dollar.

But what’s crystal clear after the passing of the GENIUS Act is that stateless money will also have to compete with state money that runs on blockchains, which will use their existing network effects to bid for supremacy in an age of stateless financial infrastructure.

Odds & Ends

• Gnosis launches v2 of Circles, a new community currency Link

• Combined metrics for tracking smart contract networks Link

• Lido proposes dual governance structure upgrade Link

• Trump memecoin holders prep for private dinner reward: ‘I’ll wear a suit’ Link

• Fluid struggles with impermanent loss after big ETH move up Link

• New draft market structure bill released by House Republicans Link

• DeFi Llama is now tracking risk curators by TVL Link

Thoughts & Prognostications

• The key neutrality of baselayer markets [Alex Grieve & Rodrigo Seira/Paradigm]

• A brief history of token launches [Mary/First Contact]

• The price of trust [Paul Frambot/Morpho]

• Crypto and the Evolution of the Capital Markets [Tuongvy Le & Austin Campbell]

• Dawn of LP profitability [Bunny Diaries]

• Why we need new blockchains [OneTrueKirk/Morpho]

• From Vanugard to Believe App, investing collapses into short form content [Lex Sokolin/Fintech Blueprint]

• The case for and against multiple current proposers [Rain & coffee]

That’s it! Feedback appreciated. Just hit reply. Had a great time writing this. Feeling optimistic about DeFi!

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO/Sky. Some of my compensation comes from MKR, so I’m financially incentivized for its success. All content is for informational purposes and is not intended as investment advice.

- Chris

The crypto user experience (UX) has historically been poor, with individuals losing funds, opting for centralized solutions, or leaving the space altogether.

There have been two key efforts to enhance DeFi UX. First, via centralized solutions that offer a white-glove version of DeFi. The DeFi Mullet is a world where onchain protocols are used, but most users rely on centralized services like Coinbase to hold their keys and control their funds (best exemplified by the Coinbase integration of DeFi lending protocol Morpho). The second is via a world that fulfills the original crypto vision, where self-custody is used by the masses and crypto wallets are ubiquitous.

DeFi power users – guilty as charged – are willing to put up with the clunkiness because they want the freedom to explore every inch of the blockchain. But a regular user doesn’t want anything to do with private keys.

Crypto wallets have long disappointed; every cycle promises a better wallet design to finally unlock mainstream adoption. This time, the hype might be warranted. Or at least, it's a last shot before the centralized services start driving onchain product development.

With the upcoming Pectra Hard Fork on Ethereum (introducing EIP-7702) and fresh revenue models (from in-app swaps to MEV flow deals), wallets are evolving from clunky key-management tools into sophisticated ‘orchestration layers’.

In this piece, we’ll explore how EIP-7702 builds on recent upgrades (i.e. EIP-4337), which wallets are best positioned to capitalize on user-friendly features, and why these changes could finally lead wallet-UX to fulfill its promise of self-custody.

Account abstraction: What the EIP-7702 brings

Efforts to enhance UX in crypto have been progressing recently. As discussed in our article titled Abstract away! The race towards interoperability, the latest significant upgrade to Ethereum (EIP-4337) went live in 2024. Now, market attention has shifted to EIP-7702, which is set to launch with the upcoming Pectra Hard Fork. In other words, an opportune moment to discuss its implications. The key one being that, with this new infrastructure in place, we’re finally equipped to build user experiences that can approach Web2 usability – while preserving the core principles of Web3.

EIP-7702 will significantly alter how Ethereum accounts function. Here are some notable improvements that will be made to existing externally owned accounts (EOAs), the default account type:

• No more token approvals: Approvals and swaps will be bundled into a one-click experience, instead of the clunky current one. Going forward, an EOA can delegate to a smart contract that executes both the approve and transferFrom steps in a single transaction. Bundling also enables more efficient execution of complex DeFi strategies without requiring individual transaction signatures.

• Gas abstraction: Users will be able to pay gas fees with any token, benefit from sponsored transactions (where an application or wallet covers gas costs), and utilize cross-chain gas payments. Currently, users must acquire ETH or SOL to send USDC – imagine having to buy Federal Reserve stock just to wire dollars.

• Account recovery: Depending on the wallet provider, recovery options such as guardians, email-based recovery, and legacy/inheritance features will be introduced. Many crypto users have lost tokens by misplacing their seed phrases.

• Less signing: Users will be able to set predefined rules for interacting with decentralized applications, providing limited access without requiring constant approvals. With EIP-7702, users can create session keys to help mitigate the associated friction by allowing users to pre-authorize interactions, reducing the need for excessive signing.

Despite the benefits, the speed at which EIP-7702-based wallets will be adopted is uncertain. One thing about wallets is that users are sticky – and products are slow to implement new features. For Dapps, this creates twice the complexity: two user journeys, two separate interfaces – almost like building two distinct products. As a result, there’s little immediate motivation for Dapps to invest in dedicated account abstraction experiences, and users have little reason to adopt smart accounts if they don’t noticeably improve their Web3 experience.

Ambire, Argent and Avocado are among the first wallets to add EIP-4337-based smart contracts. In the developer tools space, Thirdweb and Alchemy offer SDKs and APIs that make it easy for developers to embed EIP-4337-compliant smart wallets into dApps, accelerating adoption.

In 2024, EIP-4337 smart accounts saw massive growth, with over 103 million User Operations – a more than tenfold increase from 2023. (Side note: UserOps are a new user transaction type after the EIP-4337 update). However, most accounts were single-use, driven by apps offering sign-up rewards, while only a few apps, like Blocklords and OpenSocial, succeeded in driving repeat engagement. Notably, 87% of UserOps fees were covered by Paymasters, making transactions cheaper for users. Overall, EIP-4337 adoption expanded across multiple chains and use cases, especially gaming and social apps (though retaining active users remains a challenge). Still, ERC-4337 bundle transactions represented only about 1.5% of total ETH + L2 ecosystem transactions in 2024.

One critical aspect missing from the wallet-UX-upgrade movement is privacy. Vitalik Buterin recently emphasized that privacy remains an unresolved issue in crypto. Privacy-focused cryptocurrency Monero, despite its talented developer team and significant financial resources, has never achieved mainstream adoption. While the removal of Tornado Cash from the OFAC sanctions list is a positive development, the blockchain community – including the Ethereum Foundation – has yet to prioritize privacy enhancements at the Layer 1 level. In his December article, Vitalik poses some short-term privacy solutions based on wallets, some of which are being implemented in Zupass. However, to have a proper long-term solution, these improvements need to be implemented at the base chain or application layer, like the recent launch of PrivacyPools.com.

The new UX stack: AI agents

Outside of the infrastructure improvements, wallets may also see improvements with the rise of AI. Bitte.ai is pioneering AI-driven automation by embedding AI agents directly into DeFi applications. Instead of traditional point-and-click interfaces, Bitte’s system employs chat-based interactions, where AI assistants guide users, execute transactions, and manage DeFi activities.

Projects such as Bitte allow users to input natural language commands (e.g., “Swap X ETH for X USDC on chain X,” “Deposit to Spark,” “Send 0.5 ETH to Mom”). AI then determines the optimal route, constructs the transaction, and presents it for confirmation. This text-based approach improves usability by consolidating multiple DeFi protocols within a single interface. To enhance security, AI analyzes user behavior and introduces additional confirmation steps for suspicious activity.

One of the greatest benefits of the shift to AI-driven or chat-based UX is the rise of intent protocols. As we previously discussed, intents represent a user expressing their desired outcome in a transaction request. This approach removes unnecessary constraints on how the request is executed, focusing instead on the intended result. Specialized actors known as ‘solvers’ then compete to identify the most cost-efficient way to fulfill a user’s intent.

So to recap, wallets will benefit from major infrastructure upgrades on Ethereum with 7702. And with this, be able to wield AI in their never-ending battle to abstract away the complexity of blockchain transactions.

Wallets can finally make money

For most of crypto’s history, wallets were critical infrastructure with no business model. They sat closest to the user but struggled to capture value, serving as UI layers that were expensive to build and maintain, yet difficult to monetize. That's now changed. Today, wallets are quietly becoming profitable. MetaMask generates $50 million+ through in-app swaps. Phantom just raised at a $3 billion valuation with a reported $90 million in annualized revenue. Many wallets now monetize order flow or capture MEV — similar to how Robinhood monetizes retail flow in traditional finance.

Wallets have always been alluring to investors, but never sustained the investment required to make them consumer products. With a more predictable revenue stream, investors and companies will surely be more willing to dump resources into marketing and user acquisition for wallets.

Abstraction = Commoditization (and the rise of wallets)

These changes carry major business implications. Today, users must manually select a blockchain, adding unnecessary complexity. In the future, intermediaries will automate this process based on security preferences (e.g., institutional investors favoring highly secure chains) or specific user goals (e.g., choosing a chain that supports a particular DeFi protocol).

Entities that implement this selection – such as wallets and intent protocols – could become the dominant distribution channels, effectively controlling the user relationship. This shift would benefit risk managers like Gauntlet and Block Analitica, as well as risk assessment frameworks like L2Beat. This would then influence which chains and protocols AI defaults to as ‘preferred’ options. By default, assets may be allocated to safer networks such as Ethereum or Stage 1 L2s like Arbitrum and OP (as categorized by L2Beat), while advanced users could opt for riskier alternatives, including Stage 0 L2s.

However, this trend could face pushback from L1 and L2 chains, as it risks commoditizing them. L1s and L2s are among the most well-funded and deeply entrenched players in the industry, making the competitive dynamics unpredictable. Uniswap has already launched its own blockchain and later introduced a wallet, so is likely to maintain control over its user relationship. Meanwhile, we see ‘Ethereum-aligned’ wallets like MetaMask directing traffic toward Ethereum and its L2s. Yet as MetaMask adapts to market demand by integrating Solana, one must ask: will the Ethereum Foundation need to launch its own ‘Ethereum Wallet’ to defend its moat?

Who owns the end user?

Wallets now have the upgrades, business models, and AI-driven UX to push DeFi towards the mainstream. Still, there’s no guarantee of a purely self-custodial future.

On one hand, the DeFi Mullet isn’t likely to go anywhere: centralized platforms already dominate onboarding and can simply fold in onchain back-ends for those wanting ‘crypto under the hood’ without key-management hassles. On the other, EIP-7702 and ‘wallet-as-orchestrator’ models give self-custody its best shot yet. If wallet providers pair account abstraction, AI-driven interfaces, and new revenue streams (like MEV flow deals) to fund innovation, they could finally overcome the clunky UX that has deterred everyday users.

We have good reason to want self-custodial wallets to succeed. They uphold crypto’s founding principle of decentralization and prevent lock-in. If widely adopted, user-controlled wallets would mean no single platform can hold funds hostage or dictate the rules.

Whether these paths merge, coexist, or collide is still to be seen – and the deciding factors go well beyond user interfaces. As we explored in DeFi splits in two in the fall, regulations, liquidity fragmentation, and protocol governance also push DeFi in divergent directions. In the near term, we may see the DeFi Mullet thrive with newcomers seeking simplicity, while more advanced or privacy-conscious users embrace the new wave of powerful, user-friendly self-custody.

If that latter group can deliver a ‘just works’ experience to rival centralized apps, they may fulfill crypto’s original vision. But if key-management complexities remain unsolved, custodial giants will remain the face of DeFi. One way or another, wallets are now best positioned to decide which version of ‘financial freedom’ wins out – and how decentralized the future of DeFi will be.

Odds & Ends

• Aave integrates Chainlink SVR on Ethereum for liquidation MEV recapture Link

• CowAMM suffers small exploit exploit Link

• Rath Finance launches new DeFi rails for yield packaging Link

• Hydro team develops new approach to impermanent loss Link

• Compound Labs reemerges, toys with launching Compound Foundation Link

• DeFi revenues decline in March as onchain activity slows Link

• AtlasEVM introduces universal auction system for transaction execution Link

• Matcha expands to Solana with SVM and EVM chain support Link

Thoughts & Prognostications

• Bunni: How to build a leading Uniswap v4 hook [0xLukasinho/Auditless]

• Celestia's roadmap: TIA time, or is DA a commodity? [Delphi Digital]

• Markets, manipulation & the myth of decentralization [Decentralised Co]

• DeFi ecosystems analysis [Our Network]

• Unpacking decentralized training [The Knower]

• Euler v2: Six months in [Euler Finance]

• Rise of AI agents [Flashbots]

• Leveraging blockchain in a new era of antitrust [DRC & Flashbots]

That’s it! Feedback appreciated. Just hit reply. Still into this meme: open-source capitalism.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov, Zhev and Financial Content Lab. All content is for informational purposes and is not intended as investment advice.

- Chris

ps - in Denver next week!

While the prospect is unsavory for DeFi purists, there's now little doubt that crypto's greatest product – apart from stablecoins – is permissionless tokenization (and the trading of such tokens). Best case, these tokens can be considered analogous to company stocks, granting users governance rights to pilot associated products towards optimality. As it currently stands, they’re simply a means to convert attention into monetary gains.

Over the years, we've seen the evolution of various models of decentralized exchanges (DEXs) that seek to facilitate the trading of tokens. The distributed nature of blockchains reduces their ability to support conventional limit order books, as is common in centralized exchanges. This is why the AMM model is more commonly adopted for onchain trading. As blockchains have scaled and trading automated, we’ve seen a convergence between order books and AMMs to the point that they’re now (almost) indistinguishable.

Much has been learned since the days of 0x and Bancor. The speculation and frenzy of DeFi, NFTs, and memecoins has spurred newer and better exchange designs that are close to an optimum state of usability. At the core, these designs are all focused on minimizing and democratizing MEV.

Below, we zone-in on the onchain market-models development trend by examining top players in derivatives and spot trading. Specifically: Drift, Jupiter, dYdX, Hyperliquid, and Uniswap.

Based on our analysis, it seems we’re close to the end state for market design. And that the winner of this round in DeFi will be the one who knocks off TradFi.

DEX design: It all starts with the underlying chain

Before we get into the review and analysis, a quick recap on general properties and considerations when building exchanges as they relate to underlying blockchains.

The initial model for blockchains was a single data layer, which all sorts of activities can be coordinated and recorded on, evident in Bitcoin, Ethereum, and Solana's design. We refer to this model as ‘general purpose chains’, i.e., chains that aren't built to cater to any specific application category, but rather to support as many as possible.

Generally, this model faces a trilemma of tradeoffs between security, decentralization, and throughput: optimizing for any two of these goals lessens the chain's chances of achieving the third. It’s a subjective belief for spectrum-based metrics – but a widely acknowledged one – that Ethereum has prioritized security and decentralization at the cost of throughput. This is similar to Bitcoin, which is even less decentralized these days, but in contrast to Solana, which pursues security and throughput with less focus on decentralization.

So while applications can launch on Ethereum to access better security and censorship-resistance guarantees, Solana is arguably a better chain for latency-sensitive applications such as limit order book markets.

Nevertheless, it’s not news that general purpose chains are inherently limited in the amount of throughput they can offer applications building atop them. Even more so if they wish to maintain a credible level of decentralization/distribution. Furthermore, most applications may wish to retain their value rather than leaking it to the underlying chain through MEV. These are the driving ideas of the appchain approach.

Into the weeds: Appchain design

All application-specific chains (or simply, appchains) have to make design choices concerning their consensus network/mechanism, preferred virtual machine, whether to be an L1 or L2, and other optimizations. L1 appchains have the benefit of ground-up building that enables them to make improvements by customizing components, while L2 appchains are easily composable with their L1s – and potentially other L2s – making it easier to attract liquidity.

All these factors must be considered, especially for onchain exchanges, as the slightest misconfiguration in setup could lead to an incorrect liquidation, a bad order match, or any other myriad of faults that would scare away user liquidity. Perhaps unsurprisingly, most teams prefer the L1 appchain setup for customization benefits and deal with liquidity attraction as a secondary issue, rather than risk reliance on an external avenue.

The table below summarises the key characteristics across the three main approaches to choosing a chain for an application.

It’s now time to go over the architecture of the five aforementioned dominant DEXs. Given the considerations each will have had to make around design (as highlighted above) we present these DEXs categorised under each of the three main approaches.

Drift and Jupiter: The performant L1 approach

Drift

The Drift protocol is built atop Solana, allowing it to comfortably serve limit-order-book-based exchanges within certain bounds. It is an onchain decentralized exchange that settles user trades via three routes:

1. Just-in-time (JIT) liquidity auctions: Orders are submitted to a committee of market makers who compete to satisfy the order at the best possible price within a specified window.

2. Limit order book: Users specify their preferred order settlement prices during submission to a network of ‘keeper bots’. These bots each maintain an offchain index of submitted orders, which they can either:

   1. sort and match by price-time priority to other limit orders

   2. use to fill a JIT liquidity auction, or

   3. settle via the virtual AMM's reserves.

3. Virtual automated market maker (vAMM): A last-resort liquidity source for the guaranteed settlement of user trades.

Jupiter

While Jupiter is mostly used as an aggregator, it has also developed a derivatives market, offering users up to 100x leverage on a limited number of assets. Jupiter’s aggregator product enables it to have better flow and deeper liquidity reserves, which ultimately imply better order settlement prices for users.

The derivatives exchange is supported by the Jupiter Liquidity Provider pool (JLP) which acts similarly to an AMM pool to accept user assets as a liquidity backstop for orders, however the users’ orders are for derivatives, rather than spot positions.

Processing of orders occurs in a two-step process:

1. The user submits a “request” transaction of their order to the chain via the application's front end.

2. A keeper monitors the request transaction onchain and executes it using liquidity from the JLP pool.

The performant L1 approach for DEXs rests on the condition that the revenue of privileged agents (e.g. block builders, validators, proposers etc.) in the default mechanism is greater than the revenue they'd get from malicious actions such as sandwiching. Various L1s have adopted extraneous guardrails for these problems: Flashbots for Ethereum, Jito for Solana, and Skip for Cosmos. However, speedy order processing with adequate censorship-resistance is still not in sight.

In the case of Drift/Jupiter and Solana, the underlying L1's stated goal is to serve as a decentralized NASDAQ, or more recently to “increase bandwidth, reduce latency”. This means pursuing throughput and security at the expense of decentralization. As Solana's throughput increases, the hardware requirements for validators do as well, causing more validators to fall behind or shut down operations entirely.

This leaves the network in the hands of only a few validators who will inevitably begin exploring other avenues for revenue aside from the foundation's subsidy. This also means applications atop (such as Drift) will begin to leak value to such ‘unaligned’ validators, leaving users worse off due to MEV.

Nevertheless, the search for the perfect L1/L2 continues with the likes of Movement, MegaETH, Monad, and Atlas scheduled to enter the frame very soon.

dYdX and Hyperliquid: The L1 appchain approach

dYdX v4

dYdX was one of the first providers of onchain derivatives. The dYdX Team has since pivoted from an Ethereum Layer 2 chain offering, to building a standalone L1 within the Cosmos hub. V4 was launched as a CosmosBFT-based L1, to enable the protocol to take advantage of the mechanism's relatively unopinionated design specifications and customize validators’ duties to increase throughput.

The dYdX chain is populated by validator nodes (which are responsible for gossiping/executing orders and finalizing blocks) and full nodes (which pass real-time data to indexers). Therefore, the chain's p2p network is responsible for:

1. Executing received orders by matching them to each other.

2. Including matched orders in blocks and extending the chain.

3. Providing the data related to order execution to users.

Part c. is carried out collaboratively with indexers, which are read-only data endpoints optimized to serve users similarly to RPCs in Ethereum. Indexers ingest data streams from full nodes and decompose them into either onchain or offchain categories, before serving them to users or anyone else.

Having a p2p network with customizable functions enables the dYdX chain to implement a novel MEV mitigation scheme through ‘vote extensions’. Its strategy is two-fold as it:

1. Eliminates the block proposer's first-look privilege by enabling collaborative block building with other validator nodes, effectively simulating a leaderless mechanism (although execution is still entirely the proposer's duty).

2. Implements a frequent batch auction (FBA) in each block for similarly priced orders, so that ordering advantages are minimized.

Hyperliquid

Another example of the appchain approach, Hyperliquid has gained impressive traction within barely two years of launch. This was initially due to its smooth UX relative to competitors, with users hailing it as an onchain, no-KYC centralized exchange. Then came its HYPE token, which has become the new standard for fair product token launches.

The Hyperliquid L1 is a PoS chain that runs on a variant of the HotStuff consensus mechanism referred to as HyperBFT. This mechanism is optimized to enable validators to run a low latency order book that serves users at a self-reported average rate of 100k orders per second.

So far, it seems the L1 appchain approach hasn't lived up to the associated hype due to the unique issues both dYdX and Hyperliquid face. On one hand, dYdX remained steadfast to the open-source/decentralized ethos of crypto to build an L1 that is sufficiently censorship-resistance for its purposes. However, it has been criticized for its weaker performance that has caused it to lose a great portion of market share. Its token distribution model is also being called under question as the underlying cause of its continued underperformance, especially since it could easily be viewed as extractive and unfriendly to retail investors (relative to Hyperliquid's friendly airdrop and distribution model).

On the other side, Hyperliquid has gone its own route with mostly closed-source developments and a centralized model. Critics say its rapid rise to success is due to a great deal of centralization that persists at almost every level of the application. Its proponents tend to disagree, especially given its continued outperformance on almost every metric. However, these two arguments don't overlap; if anything, Hyperliquid’s continued success points to the risk profile of its users. We believe it’s a good product, just not a DeFi product. Yet.

Unichain: The L2 appchain approach

While we present Unichain as an example under the L2 appchain approach, it’s worth noting that the developing team (Uniswap) is perhaps better classified as a ‘composable stack’. This is because their products span virtually every area, from wallets to various DEX models, and now a DeFi appchain.

Uniswap started its journey as a permissionless AMM with asset prices defined by their quantities in a pool. Over the years, the initial model has been altered to meet the demands of an ever-evolving user base, as summarized in the table below:

Uniswap v4

The latest iteration comes with a great deal of improvements, the most notable being hooks and the singleton architecture:

1. Hooks are subsidiary contracts that can be called at a specific point during a user's interaction with a liquidity pool, in order to trigger a pre-designated action.

2. The singleton architecture is an optimization for saving gas, but it also enables what is referred to as ‘flash accounting’. This system enables Uniswap v4 to transfer assets only on net balances, so that swaps involving multiple liquidity pools will require less onchain storage updates and consequently be more efficient.

Together these features shift Uniswap v4 towards the modular design principle for lending markets we discussed previously. As such, it’s no longer a simple product but a platform, to which complexity can be safely introduced by developers without loss of composability.

UniswapX

Aside from its primary product, the Uniswap team has also created an RFQ market, UniswapX. This is essentially an intent-powered market wherein users can define their preferred execution conditions for a trade, while fillers compete in an auction to satisfy user preferences.

While seemingly different, both Uniswap v4 and UniswapX are actually complementary, as has been laid out by the team. The introduction of hooks and arbitrary fee values in Uniswap v4 leads to greater liquidity fragmentation across unique pools, leading to higher routing complexity that translates to higher transaction fees for the user. While the Uniswap auto-router is optimized to solve this problem, there is no guarantee that its selected route for a user's transaction is the most optimal; thus users pay more with no guaranteed outcomes.

This problem is being addressed by UniswapX, which allows users to set tight bounds on their expectations while offsetting execution to experienced fillers, who have access to more information and inventory, and compete to satisfy users for a fee. Cowswap is coming at this from the other direction, starting first with an intents-based aggregator and then designing a MEV-capturing AMM.

Unichain

The Uniswap team has also announced they are building out a new rollup tailored for DeFi applications called Unichain. While this came as a surprise to some, it makes sense that one of the biggest drivers of order flow would want to control it better, especially since better flow control implies better MEV mitigation (among other things).

Moreso, the remodeling of Uniswap v4 as a baseplate will inevitably drive the need for a more performant base layer, one which can easily support hooks’ features, especially the necessary throughput. For example, in the case of the speed necessary for onchain limit orders. Uniswap v3 already had the simplest form of limit order books with its ticks, so v4's hooks will inevitably perfect this, and then require more infrastructure support.

Unichain can easily satisfy hooks’ latency demands with its ‘flashblocks’ (essentially glorified pre-confirmations), while reducing the toxic flow users are exposed to due to ordering via its sequencer-builder separation model.

CEX and TradFi: Final bosses

While these DEX projects are all competing with each other, they’re really after CEXs and King Binance, where most derivative and spot trading takes place. It’s telling that there hasn’t been a new successful upstart CEX this cycle. There’s no FTX trying to challenge Binance. In fact, it’s Hyperliquid that has finally eaten into Binance’s overwhelming lead. And the latter certainly feels threatened, making (in)direct shots at Hyperliquid on X.

Hyperliquid feels very much of the moment. It is rightly criticized for its extremely centralized model, but if we take a step back, we can see that it represents an evolution of new exchanges launching with more and more crypto-native financial infrastructure first. Coinbase was a CEX, but then Binance launched with a token from the start. Now Coinbase has its own L2 and while Binance.com is dominant, BSC is arguably one of the big three smart contract blockchains, along with Solana and Ethereum.

In fact, all new innovation in crypto exchanges these days is coming from DEXs and in DeFi. CEXs gave us the perpetual derivative – a genuine financial innovation – but DEXs are quicker to list tokens, unlock new yield opportunities, new pooled lending, and most importantly, are driving the RWA push. Coinbase and Binance are not trying to innovate TradFi with their CEXes. They’ve bet on Base and BSC to do that.

The key question is whether the DEX that slays TradFi will be one that specializes in strong distribution and on & offramps, like Binance, Coinbase or Hyperliquid, or from one where the tech shines first (be it an L1/2 app chain or high-performant general purpose chain). Our bet is on the infrastructure wagging the distribution, eventually.

Odds & Ends

• North Korea hacks Bybit for $1.5bn Link

• Coinbase says SEC has agreed to drop enforcement case Link

• Hummingbot releases v2.3 Link

• Vitalik proposes higher gas limits for Ethereum L1 Link

• Berachain: A story told through charts Link

• Ethereum Foundation deploys 45,000 ETH into DeFi Link

• Overview of new regulatory efforts in US congress Link

Thoughts & Prognostications

• The need for clarity on decentralized stablecoins in crypto legislation [Luca Prosperi/M^0]

• The road towards a distributed encrypted mempool on Ethereum [Shutter Network]

• The Javier Milei Libra scandal and financial nihilism [Lex Sokolin/Generative Ventures]

• Exploring AMMs [InfectedCrypto]

• Choices around sequencer architectures will be influenced by the economics [Chris Burniske/Placeholder]

• Pectra: Ethereum’s next major upgrade [Tanay Ved/CoinMetrics]

• The emerging liquidity fragmentation crisis on Ethereum [Peteris Erins/Auditless Research]

  ---

  That’s it! Feedback appreciated. Just hit reply. I’m into a new meme: open-source capitalism.

  Dose of DeFi is written by Chris Powers, with help from Denis Suslov, Zhev and Financial Content Lab. All content is for informational purposes and is not intended as investment advice.

Donald Trump’s memecoin launch on the eve of his inauguration was nothing short of sensational. Within days, $TRUMP soared to an $14 billion market cap, with Melania Trump’s $1.6 billion memecoin following closely behind. This spectacle dominated headlines, sparking debates across the crypto ecosystem. Is this a pivotal moment that legitimizes the industry? Or more a glaring reminder of its speculative underbelly?

Crypto is winning — there’s no doubt about that anymore. Even a laypersons’ quick look at the current price of Bitcoin would solidify this. But a harder question lies ahead: what kind of crypto vision with the largest cultural impact will emerge? Will it be defined by the financial nihilism of $TRUMP, Bitcoin’s digital gold, Ethereum’s decentralized ethos, Solana’s global permissionless NASDAQ vision, or something altogether different? And perhaps most importantly, where will DeFi fit in? With $TRUMP, BTC, SOL and ETH seemingly at inflection points (more below), we expect to see DeFi playing a far more foundational role come the next presidential inauguration in 2029.

The $TRUMP phenomenon

Unprecedented launch

The launch of $TRUMP was a masterclass in leveraging attention and liquidity. Within 24 hours, the token’s price skyrocketed from $7 to $75, before stabilizing at $38. It became the 21st most valuable cryptocurrency overnight. By comparison, Melania’s memecoin, despite its strong debut, struggled to maintain momentum. Together, the Trump tokens epitomize the unpredictable nature of memecoin economics.

$TRUMP was unprecedented in scale. Previous memecoins like DOGE and PEPE took years to reach their current size but $TRUMP reached this in just hours, which leads many to wonder if this is just the beginning of tokens directly tied to public figures.

Memecoins and financial nihilism

Critics have labeled $TRUMP a symbol of financial nihilism — a “get-rich-quick” scheme that undermines the credibility of crypto. Some more neutrally argue that $TRUMP represents the duality of crypto: it is both a tool for empowerment and a magnet for greed. While some celebrate $TRUMP as proof of crypto’s mainstream success, others warn that it sets a dangerous precedent, trivializing the space as a playground for speculation.

This spectacle has also reignited debates about the future of tokenization and its role in broader economic systems. As Noah Smith argues, memecoins like $TRUMP and $MELANIA embody the "slush fund" aspect of crypto: assets that allow individuals to bypass TradFi’s constraints while simultaneously fueling speculation. Whether this is a bug or a feature, you’d have to ask the Trumps. But we don’t think this is the post-modern future that cypherpunks and decentralization zealots signed up for.

Bitcoin: Sitting pretty, but what’s next?

Bitcoin is in a great spot, particularly with the pardon and release of Silk Road founder Ross Ulbricht, a move celebrated across the Bitcoin community. But where can it go from here? There’s the oft-touted Bitcoin Reserve, an idea that seems obviously bullish on the surface. Despite this, prominent Bitcoin enthusiasts have raised concerns on the reserve idea, questioning whether this could open the door for government co-option of Bitcoin’s inherently decentralized ethos. And that it would be silly for the U.S. government to prioritize a rival to the dollar.

This then begs the question: is Bitcoin chasing another level of adoption? As digital gold, it’s already had institutional buy-in, which is likely to serve as a long-term price driver. But Bitcoin once aimed higher. Hyperbitcoinization — a term describing the world running on Bitcoin as its primary currency — was the ultimate goal. Are we still on that path? At this point, it seems unlikely. Bitcoin’s trajectory is more akin to a widely accessible inflation hedge rather than a revolutionary societal currency.

Hyperbitcoinization, by definition, seeks a world where Bitcoin replaces fiat entirely. But to achieve this, Bitcoin must offer more than just price stability and ease of access, and its cultural and societal ambitions have largely taken a backseat. Whether it can reclaim that aspirational vision is an open question. But for now, Bitcoin remains a highly functional, if somewhat static, cryptocurrency.

Ethereum: Reflexive spiral downward

Things are less rosy for Ethereum. Prices are falling, and sentiment within the community is increasingly negative, as evidenced by the growing trend of users abandoning their ‘.eth’ handles. Criticism of Ethereum’s leadership, which has been accused of being overly academic and disconnected from real-world applications, has grown louder. The top dogs are only now beginning to acknowledge the need for change, with Vitalik Buterin entering "wartime mode" (or just coincidentally changing his profile pic, Milady). Plus, a huge new ETH research post on native rollups was just posted by Justin Drake.

Ethereum still holds onto its ambition of building resilient, decentralized systems. However, its biggest hurdle is the lack of effective decentralized coordination. While Ethereum has been a foundational layer for DeFi, its inability to foster thriving DAOs has hindered its progress. This gap in decentralized governance could be what’s missing from Ethereum’s equation. DAOs, as a framework for decentralized decision-making, could unlock Ethereum’s next phase of growth. (Side author note: this is why I’ve spent the last few years focused on the DAO space, and now on Powerhouse).

DeFi remains Ethereum’s most successful application to date. Platforms that allow users to generate yield, trade assets, and interact with permissionless financial systems underscore its utility. However, Ethereum’s fragmented liquidity across Layer 2 solutions and a lack of cohesive governance risk its future dominance. The network must move beyond its current state and create a unified, scalable vision for decentralized applications. Just today, Vitalik dropped another banger: Scaling Ethereum L1 and L2s in 2025 and beyond.

Solana: Thriving

Solana’s role as the blockchain for meme coins and high-performance applications continues to expand. While it’s true that running a Solana node requires significant compute power — which has raised concerns about potential government coercion — the network has demonstrated its value through memecoin trading. $TRUMP’s launch showcased Solana’s ability to gather massive amounts of liquidity on a single chain, although it did struggle to maintain uptime during the frenzy.

The area where Solana is most ahead is in token launches. Ethereum used to be clogged for NFT mints, but this got too expensive. Some L2s stepped in but none has been more successful than Solana. It’s clear now that if you want to “drop” a token, Solana has the best distribution even if it gets congested.

But can Solana leverage its momentum to expand beyond memecoins and secure a foothold in DeFi? While Solana is the favorite of new crypto czar David Sacks and is making major strides in its MEV supply chain and building high-performant DeFi systems, it’s noticeably behind in RWAs. Plus, it’s not clear that its advantages over Ethereum will carry over to a more regulated DeFi world. Relatedly, token sales are primed to make a comeback (make ICOs great again!). Will that look more like the burgeoning RWA market or the memecoin market?

The regulatory frontier and stablecoins

The regulatory environment for crypto is entering a new phase with Trump in office, and perhaps more importantly, a more friendly Capitol Hill. The establishment of the SEC’s new Crypto Task Force, led by Commissioner Hester Peirce, marks a significant milestone. Tasked with creating a comprehensive and clear regulatory framework, the initiative aims to move beyond the SEC’s reliance on enforcement actions to regulate retroactively. Instead, the Task Force seeks to draw clear lines, provide realistic paths to compliance, and foster innovation within legal bounds. Meanwhile, Trump’s Strengthening American Leadership in Digital Financial Technology Executive Order sets an overall tone that is breeding optimism across crypto.

Stablecoins are likely to be what takes up most of the oxygen in Congress in 2025. There will be new rulemaking at the SEC and an abandonment of the regulation-by-enforcement approach, but legislation is the only way to make lasting change. Crypto did extremely well in the 2024 House and Senate races, so not only will there be more crypto allies, a Republican majority, but Democrats will be itching to showcase their moderate credentials and commonsense stablecoin legislation is a great opportunity to do that. Broader changes to securities law by congress may be hard to come by. But stablecoins and a friendlier SEC may be enough to create clearer pathways for token issuance, fostering broader adoption. Indeed, BlackRock CEO, Larry Fink, renewed his calls for tokenized bonds and stocks just this week.

Tokenization: The universal ‘aha moment’

We won’t make any price predictions for $TRUMP but we do think the cultural effect will be longstanding. Not because everyone will have a meme coin but because it will be the ‘aha moment’ for many digital asset investors around tokenization’s power to transform traditional assets into liquid, tradeable digital tokens. From equity-like structures to loyalty rewards, tokenization bridges gaps between ownership and utility in ways traditional systems never could. This extends beyond financial assets: think art, real estate, or memberships.

It’s not just new assets, but how tokens can act as both quasi-equity and functional tools for engagement with users/equity holders. Who knows what perks Trump will offer to $TRUMP holders, but the model could be replicated by others. Of course, for this to work, you need liquidity in these tokens, which is where DeFi needs to shine.

This isn’t a technical breakthrough; it’s a philosophical shift spurred by the launch of $TRUMP. The ability to tokenize anything fundamentally changes how people interact with assets, fostering transparency, reducing barriers to entry, and creating liquidity where previously none existed. It’s not just a technology, it’s a new way of thinking about ownership and value. Of course, this means there will be A LOT of shit coins. Watch out.

Closing thoughts: The road ahead

The $TRUMP phenomenon is a microcosm of crypto’s broader narrative: a mix of promise, peril, and potential. As memecoins capture headlines, the industry must grapple with the long-term implications. More specifically, on whether the future of crypto will be defined by speculative assets, or a return to its roots of decentralization and striving to create more efficient systems and real-world utility.

The road ahead is as much about preserving the values that define the industry as it is about embracing innovation. Censorship resistant money, stablecoins, and prediction markets have emerged as undeniable crypto use cases. DeFi, with its emphasis on transparency and programmability, offers a path forward and a foot in the door for decentralization.

To us, a world of systems, networks and institutions without centralized control is still the ultimate end game for crypto. And that means it’s not just about more efficient technology. It’s about a change in culture. Always has been. And we just don’t believe the $TRUMP memecoin represents such a change. Rather, the cultural shift is still about decentralization as it moves from the driving technological concept in crypto over the last 15 years to the most important political concept of the 21st century.

Odds & Ends

• Circle acquires Hashnote and USYC Link

• Erik Voorhees takes down Jack Mallers a notch or two Link

• 1kx, ParaFi and Framework lose appeal in legal battle against Curve founder Link

• Crypto committee chairs for the 119th Congress Link

• Block Analitica’s Sphere dashboard gives aggregated DeFi borrow rates Link

• Liquity v2 launches on Ethereum Link

• Uniswap plans release v4 this week and launch next week Link

• Recording from Columbia CryptoEconomics Working Session Link

Thoughts & Prognostications

• Stablecoins are non-fungible, bank deposits are fungible [JP Koning/Moneyness]

• How to price USD0++ [Jalah]

• Usual Money: A valuation framework [Luca Prosperi/Dirt Roads]

• The Solana thesis: Internet capital markets [Kyle Samani/Multicoin]

• DeFi’s hidden choreography: Decoding 2024’s most strategic transactions [EigenPhi]

• Post-election thoughts arbitrage trading in prediction markets [Joshua Gunn/Aori]

• An intro to blockchain mechanism math, terminology and hieroglyphics for deeply casual people who want to sound smart when discussing white papers with their peers [Alex Watts/Fast Lane]

That’s it! Feedback appreciated. Just hit reply. January is cold and there is no sun. Boo. I’m in Denver next month if you’re around, get in touch!

*It’s surprisingly hard to get the AI image generators to do things with Trump in the prompt…

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO/Sky. Some of my compensation comes from MKR, so I’m financially incentivized for its success. All content is for informational purposes and is not intended as investment advice.

2024 was a year where growth and success just weren’t fast or big enough. We’ve built infrastructure, products, grown TVL and even had number go up, so where’s the mainstream impact to silence the haters? It’s already there if you squint when it comes to stablecoins, prediction markets, and non-sovereign money. From here, it then becomes a question of adoption speed and the emergence of some new version of Moore’s law relating to financial activity moving onto blockchains.

DeFi is almost ready to scale to a billion users. The chain is almost fast enough. Transactions are almost cheap enough. And we almost have enough regulatory clarity to onboard institutional and retail investors. It all feels tantalizingly close, because we can see the progress: from the depths of the bear market in 2022, and from the euphoria postDeFi Summer in 2020. At both points,this DeFi thing already felt like the future (to us).

Anyway, without further ado, let’s get to the top-five memes and top-five charts of 2024:

Top-five memes of 2024

1. Bitcoin

What a year for Bitcoin, starting in January with Bitcoin ETFs, all the way through the election and past $100k. To say ‘Bitcoin has arrived’ is a bit clichéd, but there is a new level of mainstream acceptance as a macro asset that doesn’t feel like it’s going away. Bitcoin, in the words of Fed Chair Jay Powell, is “just like gold, only digital”, which makes it the ultimate meme coin.

The Bitcoin community has also largely abandoned the Bitcoin blockchain, solidifying around the meme of a fixed supply of BTC the asset. Last year and early this year, there was talk of Bitcoin Season 2, a route for stablecoins and DeFi apps to finally come to Bitcoin via L2s. Progress so far has been minimal, and most bitcoiners are instead talking about BTC the asset in a US Strategic Reserve. Maybe the scare from Google’s Willow quantum computing chip will rekindle interest in the underlying blockchain, but it’s hard to argue with the product-market-fit of Bitcoin – especially after its stellar 2024 and its seemingly inevitable path towards digital gold and non-sovereign money.

Bitcoin going up is undoubtedly good for the rest of crypto. Whether or not alt season takes off, the overall mood towards crypto experimentation in TradFi is much brighter when Bitcoin shines. At an individual level, after holding BTC for a while and it going up, the desire is to try more. And DeFi-enabled products are adjacent-possible for the crypto curious.

Reflect: After twelve years of writing about bitcoin, here's how my thinking has changed [JP Koning/Moneyness]

2. Crypto politicians

This was also the year that crypto truly arrived in American politics, with an extremely successful 2024 campaign season up and down the ballot for crypto candidates. There’s Trump at the top, who has wrapped himself in crypto from NFTs to DeFi forks, and of course with his speech in Nashville at the Bitcoin Conference, where he called for the US to become the “crypto capital of the planet”. And since the election, things continue to look up for crypto in Washington. Even this week, with the apparent cancellation of an anti-crypto Democratic SEC commissioner in the lame-duck session.

It’s hard to describe just how big of a turnaround this has been from a year ago, or even six months ago. Back then, the SEC was actively suing Coinbase and sending wells notices to every DeFi project it could get a US address for.

As we approach year-end, Jake Chervinsky of Variant is calling for the crypto industry to rethink their legislative preferences, now that it is negotiating from a position of strength. We’ve been overly optimistic that crypto legislation would get done but Republicans are eager to accomplish something that doesn’t cost any money and some Democrats will be eager to express bipartisan bonafides.

What could this mean for DeFi? Check out DeFi splits in two for our take or check out CoinCenter’s 2025 Top policy priorities for 2025.

3. Parasitic L2s

The success of the ‘Parasitic L2s’ meme demonstrates the power of the meme. It doesn’t really matter if it’s true or not, because the message is clear. In response, the L2s have wrapped themselves in the Ethereum flag, espousing their allegiances. This social layer is an underrated market enforcement mechanism for decentralization on Ethereum, but the bigger uncertainty is on the future Ethereum roadmap. Solana’s success has shaken many in the community and questioned Ethereum’s commitment to a rollup-centric roadmap. Many claim that L2s are earning fees that should go to ETH holders, while the Ethereum L1 is abandoning its core use case (DeFi) to support L2 development.

Echoes of this conversation have dominated the discourse for why ETH has trailed BTC and SOL over the last twelve months. L2s are parasitic, while the shared state of Solana makes it easy for builders to launch their app without conducting a market research study on which Ethereum L2 to pick.

Parasitic L2s were one of many critiques hurled at Ethereum this year from an eclectic class of folks who have either migrated to Solana (Max) or who straddle multiple communities (Jon Charbonneau of DBA). These critiques have had three effects so far. First they’ve provoked an autoimmune response from some old Ethereum community members who are pushing back with the Ethereum concernoooorrrrs meme. Second, L2s are draping themselves in the flag of Ethereum. And lastly, core devs have “publicly voiced support for scaling the L1” in the last month.

So the memes (and the criticism) work?

Explore more: Abstract away! The race towards interoperability [Dose of DeFi]

4. Trading memes power Solana’s rise

During a gold rush, sell picks and shovels. And it certainly rings true for Pump.fun, the first and leading meme coin launch platform, as it has generated $180 million in revenue since its launch in March. The numbers involved are a sign of the times, with the top-three tokens launched on the platform boasting impressive market capitalizations: $1.2 billion for Peanut the Squirrel, $800 million for Goatseus Maximus, and $640 million for a third token (whose name we will not disclose here).

We see the irony in one of the year's top memes being about trading memes. Regardless, this does tell a bigger story about the rise of Solana, which has been the center of the memecoin frenzy that has driven usage metrics that surpass Ethereum. Solana has arrived in its own way in 2024, commanding enough of the market that it’s presenting growth opportunities for Ethereum DeFi projects. For example, in November, Sky launched its new USDS stablecoin on Solana and one wonders when Aave will follow (although rewriting a codebase is a bit harder than minting some tokens). Solana is also carving its own path and approach to MEV that still needs work, but luckily, Ethereum exile Max Resnick is taking his talents to Solana to help. At the moment, Solana’s ecosystem is generating more MEV. This may be because there’s more fat, yet the sizable opportunity ultimately means more people are willing to fire up Solana bots to play in the game, bringing more liquidity to the system overall. Jito has been a massive success and arguably the second most important MEV project after Flashbots.

Back to the memes and their trading. We aren’t very nihilistic here at Dose, nor are we judgmental, so we’ll keep our comments brief. Crypto financializes everything. The ability to create and distribute a token is probably crypto’s biggest unlock; it will just be used in a multitude of different ways. But yeah, maybe don’t buy or sell memecoins?

Explore MEV on Solana more with Solana’s fee-market fork in the road from April in Dose of DeFi

5. ASS (Application-specific sequencing)

Application Specific Sequencing (ASS) is a deliberate provocation. Its creators are using meme warfare to try and manifest into existence an idea around MEV architecture. And it’s largely worked? Well, for an admittedly small audience (MEV researchers). Atlas and Sorella Labs are leading the meme charge. Both have products built around the idea that applications can dictate the economic distribution of its MEV supply chain. This gives them the ability to minimize MEV for users, but also determine who can extract and how.

Building ASS applications is challenging because any attempt to bypass validators — and thus cut them out of the value chain — risks censorship from those same validators (or really, builders). In essence, today’s blockchains aren’t censorship-resistant enough to fully support ASS architectures within their existing frameworks. That’s why projects seeking to control their sequencing have resorted to launching new chains or L2s, just as Uniswap did this year. Still, new proposals like Braid, Focil, Monad, and SUAVE show promise in making ASS deployments feasible. Meanwhile, Flashbots continues pushing MEV research forward, championing its own vital meme: TEEs

Overall, it was a bit of a down year for MEV meme-wise, after dominating the dialogue over the past few years and arguably still “THE problem to solve in crypto”.

A deeper look: Order flow auctions’ bumpy road to credibility

Top Charts of 2024

Note: charts are interactive online!

1. The return of high yields

Yield is the lifeblood of DeFi. Double digit returns have always been the reason for major inflows. 2023 was a particularly tough year for DeFi yields because regular bank savings rates were higher than stablecoin DeFi deposit rates. That changed in 2024 as the market roared. This time the double-digit yield was not from random governance tokens, but from someone on the other side paying interest. We are seeing billions of dollars of borrowing at 15%+ returns all across DeFi the last month.

Why? Ethena's USDe is capturing the yield offered by the basis trade on centralized exchanges, where bullish crypto investors are willing to go long BTC & ETH even with high interest on their margin loans. This can generate 20%+ APY staking USDe for extra bullish times, which is why its market cap has exploded to almost $6bn from nothing a year ago. The yield from staking USDe (sUSDe) is lifting rates across DeFi as it represents pretty good collateral (backed simultaneously by a long and short position in ETH or BTC) to borrow against. So, smart investors have looped their sUSDe to borrow USDC, USDT or another stablecoin and then purchase USDe and stake and repeat. It makes sense to keep looping as long as the borrow rate is below the yield you can get from staking USDe. In reality, the only thing stopping more looping is because the sUSDe supply caps have been hit on Aave, Pendle and elsewhere.

Ethena's success is one of the biggest stories of the year and one wonders how many more eye balls come to DeFi if deposit rates stay double digit for much longer.

2. Stablecoins

Stablecoins have become massive crypto businesses. There’s roughly $200b in total industry market cap backed by primarily dollar assets earning an average of 4%+ yield. That’s almost $10b in interest that’s being earned by companies or distributed to DeFi investors via tokenized yield.

Tether is an amazingly efficient company, with less than 200 employees and annualized revenue of $6b+. It’s almost double its 2022 market high in circulating supply. USDC, meanwhile, is growing but no longer seems in the same league as Tether. There is pretty clear movement away from USDC towards Tether after the Silicon Valley Bank collapse in March 2023. Sky* has remained relevant but has not taken off. Its rebrand to Sky & USDS is targeted at mass retail but they haven’t gotten there yet. USDe from Ethena is not really a stablecoin (synthetic dollar is what they prefer), but its size ($6b) and much higher yield opportunities mean its going to be relevant to the stablecoin conversation regardless of what you call it.

3. Lending market share

Lending has always been DeFi’s most interesting and competitive market. Assets borrowed in DeFi have grown from $7.6b a year ago to over $22bn. The chart above shows the market share of the total amount borrowed in DeFi (a harder metric to game than TVL) with lots of changes from a year ago.

For starters, Compound and Spark/MakerDAO/Sky saw their market share decline precipitously. For Sky, this was not such a bad thing as the decrease in total amount borrowed reflects a transition to RWAs, which have better margins and are more consistent. Compound, meanwhile, bled assets as it continued to suffer from a vacuum of leadership. Morpho now has more assets borrowed than Compound, marking a huge shift in the lending market towards a modular world. Fluid, from DeFi OG InstaDapp, also burst onto the scene in 2024. And then there’s Aave, which started the year with a public split with Gauntlet to go along with tussles with Morpho, Ethena and Polygon (okay maybe it’s just Marc), but despite all of that, assets borrowed on Aave has outpaced the rest of the industry and gives Aave the most lending market share since MakerDAO in 2020. What’s more, their token price has also been on a tear, leading all DeFi tokens with a 300% return in 2024. But will their success continue or is the growth of Morpho an indicator that future growth will be in modular systems?

As we said, lending is the most interesting and competitive DeFi market

4. Private order flow %

The public mempool is not dead yet, but there has been a consistent shift away from it as users look for MEV protection from private mempools. Using the public mempool for a trade transaction is simply worse execution than sending it to a dedicated network of solvers who promise to return most of the value they could extract if they traded against you. We started the year where 10% of transactions were private and that shifted up to 30% because Metamask rolled out “Smart Transactions” to combat “frontrunning”.

MEV has been pushing transaction sequencing outside of Ethereum for several years and it looks like that is set to continue as wallets, users and applications become more MEV aware. Are more private transactions bad for Ethereum? It’s unclear but it surely changes the market structure and centralization risks.

5. Polymarket

Polymarket was such a huge story for crypto this year. Its penetration into the cultural zeitgeist makes it three successful use-cases for crypto: non-sovereign digital money, stablecoins and prediction markets. This will make skeptics more open to other crypto products and also be a funnel into other crypto products from Polymarket users.

The Election put Polymarket on the map, but will it have staying power the four years in between elections? The chart above is encouraging, showing Polymarket with over $1bn in non-election bets in December so far, double what it saw in August & September.

That’s it! Feedback appreciated. Just hit reply. I enjoyed writing this as a reflection on the year. Hope everyone has a lovely holiday season.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. *I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO/Sky. Some of my compensation comes from MKR, so I’m financially incentivized for its success. All content is for informational purposes and is not intended as investment advice.

The adoption of zero-knowledge (ZK) rollups – once hailed as the next big thing in blockchain scalability – has been surprisingly slow. Despite their technological promise and recent major leaps in privacy and scalability, implementation into everyday blockchain use has been slow. Below, we discuss reasons behind the slower uptake of ZK rollups compared to optimistic rollups, and ponder how the technology will integrate into DeFi and blockchains in the future.

The cave analogy: Understanding ZK basics

Zero-knowledge proofs allow someone to prove a claim without revealing the underlying information. While ZK involves some of the most advanced math in crypto, we can grasp the basics through analogies and high-school concepts.

The setup

Imagine a cave with two paths – path A and path B – that eventually meet at a locked door inside. The door can only be unlocked with a secret code. You claim to know this code.

• Prover (you): Knows the secret code to unlock the door.

• Verifier: Wants to confirm that you know the code without learning it themselves.

The process:

1. You and the verifier stand outside the cave. You enter the cave, choosing either path A or B. The verifier stands outside and doesn’t see which path you take.

2. Challenge: Once you’re inside, the verifier shouts which path they want you to come out from.

3. Response: If you take path A and the verifier asks you to come out from path A, you simply exit from path A. If you take path A and the verifier asks you to come out from path B, you must use the secret code to unlock the door, walk through it, and then exit from path B.

This process is repeated multiple times. If you consistently exit through the correct path, the verifier gains confidence that you know the code. Importantly, the verifier never learns the code itself – hence, ‘zero-knowledge’.

The way the cave is built in this example (a door with a secret code and how it can influence the outcome) is very similar to how ZK systems are built.

Back to school: How do ZKs actually work?

ZK proofs can further be explained in simple terms, relying on mathematics, particularly using polynomials. These algebraic expressions – formed by the sum of any number of terms of the form cx^k – are key to how ZK systems work. You may remember these from high school math:

x + 1

x²

X³+3x²+2x+1

345x³³⁵+221x³³⁴+115x³³³+...+65x+44

The important thing to remember as we try to understand ZKs is that polynomials can contain an unbounded number of terms. We can then understand that a single equation involving a number of polynomials can represent an infinite number of equations (read: relationships) between numbers. For example, consider the equation that includes polynomials A, B, C, and variable x: A(x) + B(x) = C(x). If this equation is true, then it’s also true that:

A(0) + B(0) = C(0)

A(1) + B(1) = C(1)

A(2) + B(2) = C(2)

A(3) + B(3) = C(3)

…

If some equation involving some polynomials solves for a randomly selected x, then it almost certainly holds true for the polynomials as a whole. So, when we look back to the cave example above, the person (verifier) can exit through the randomly required path multiple times, which is the same as an equation involving polynomials solving for a randomly required x multiple times. This in a succinct way proves that the prover holds the full amount of information, whether it’s a secret code as in the cave example, or relating to all transactions in an L2 block. For more of a detailed explanation, see this Vitalik’s article or Starknet founder Eli Ben-Sasson’s video.

These proofs are computationally expensive to create, but fast to verify. Here, it’s important to know that even though ZK is about zero leaked information, it is its speed property, and not privacy, that underpins rollups. Mathematicians and cryptographers are working on further reducing verification times.

ZK proofs

ZK rollups aggregate transactions offchain into batches, generating proofs (zk-SNARK or zk-STARK) that are used to validate batches of transactions. The aggregated proof, along with a minimal amount of data required to reconstruct the state, is submitted to the main chain (e.g., Ethereum). This proof is used to confirm that the offchain transactions were processed correctly. The main chain only needs to verify this small proof rather than re-executing every transaction, which significantly reduces the computational load and enhances scalability.

ZK proofs ensure that rollup operators cannot cheat or commit fraud without being detected. The proof guarantees that the state transitions and computations are correct according to the protocol. As long as the proof is valid, the results are accepted as correct, maintaining the integrity and security of the rollup.

Because ZK proofs are succinct and don’t require full transaction data to be stored or processed onchain, ZK rollups can handle a higher throughput of transactions compared to processing them directly on the main chain. This scalability is achieved without compromising the security guarantees provided by the main chain.

Types of zkEVMs

ZK technology cannot be used to verify any computational problem directly. Rather, the problem has to be converted into the right ‘form’. Normal program logic has to be represented by a group of polynomials.

You can imagine how complicated this is, and rewriting EVM (or any virtual machine) to be able to operate via these polynomials is as complex as it sounds. That’s why creating a fully-EVM compatible zkEVM is hard, and why there’s such a wide range of zkEVM implementations happening (from fully Ethereum-equivalent to high-level-language equivalent ones). This affects applicability of industry infrastructure and the extent to which smart contract code of existing projects has to be changed to run on the chain.

Why are ZKs underperforming optimistic rollups?

So, now that we understand the basics of ZK technology, it’s time to address the conundrum of ZK rollups’ lackluster takeup.

We’ve seen some successful implementations: namely, ZKsync, Scroll, Starknet. It marks a success in itself that these function at this point, with no significant hacks so far (only a couple of halts). However, optimistic rollups are still dominating ZKs, the latter holding approximately 10% of the market only by TVL and user activity. L2beat.com’s proprietary statistic, scaling factor for ZK rollups, is 2.3 times over the last 90 days. This means that only two times more transactions are settled by Ethereum thanks to these rollups. This figure is 9 times for optimistic rollups, which means optimistic rollups are being more widely used as the preferred Ethereum scaling solution.

So what is really holding ZK rollups back? We see three possible explanations:

1. Untested tech

The tech behind ZKs simply hasn't garnered as much trust as optimistic rollups. It seems tech-savvy whales are yet to move their capital to ZK rollups. Most likely, they’re waiting for ZK rollups to be battle tested and have all their bugs fixed. For example, since verifiers (that are not yet open-sourced) form such a major part of the stack, the industry can’t check and verify them.

This illustrates how making the zkEVM (or any zkVM) provable is not a straightforward task. Veridise, a security audit firm specializing in ZK audits, claims that ZK security is simply more challenging. The firm's ZK audits revealed a twice-increased chance of critical severity bugs compared to the rest of their audits. This is understandable given the complexity and rapid development of ZK technology.

Plus, the smart contracts for Type 4s (Starknet & zkSync) are not just forks of existing Solidity contracts, so they have to be written from scratch, and investors have less trust in them. Battle-tested contracts are so sought after that Starknet, initially a Type-4 ZK rollup, is adding an EVM implementation Kakarot to become a Type-3 rollup (which will support existing Solidity contracts). Full EVM and Solidity support seems to be crucial for developer adoption.

2. User stickiness

Optimistic L2s were launched three to four years ahead of their ZK competitors. And once users moved to these solutions they stayed, without seeing significant incentives to move to new ZK chains. Fees on optimistic rollups have stayed at the same level as ZKs, even though the promised fee reduction by ZKs is 10 times that of optimistic rollups. ZK rollups still lack the mass adoption needed to bring millions of users and billions of transactions to truly demonstrate their scaling advantage.

3. Tokenomics and economics

Also to be noted is that relative market capitalization of these chains is surprisingly low (zkSync, Taiko and Scroll are ranked 126, 342 and 344 on CoinMarketCap), largely explained by bad tokenomics. Low float with high FDV means tokens still have to undergo years of unlocks, deterring investors from buying them. Token launch timing hasn’t been great either; altcoins are now performing notoriously worse than in previous cycles, so ZK rollups have been launched to less welcoming markets than optimistic rollups one cycle ago. Indeed, since our piece on ZK rollups a year and a half ago, zkSync and Starknet have launched their tokens. Smaller market caps mean smaller treasuries to provide incentives to developers and users, so there is less liquidity and activity.

The future of ZK technology: Still bright

ZK technology has the potential to scale blockchains by increasing transaction throughput by several orders of magnitude, potentially up to thousands of transactions per second, depending on implementation.

The Ethereum Foundation plans to make changes to Ethereum to support ZK rollups to further improve efficiency. In the long run, ZK will actually become a major part of the L1 itself. Ethereum node clients are expected to begin experimenting with ZKs to verify Ethereum block execution on L1. We anticipate a gradual shift from clients validating Ethereum L1 blocks through direct re-execution, to most clients relying on ZK proofs for verification.

So far, ZK implementations have not introduced enough differentiation. The fees are almost the same, which doesn’t do enough to lure users away from optimistic rollups. A new technology cannot be just marginally better to convince users to switch: it must be orders of magnitude better. In the long term, once there are 100 times or 1,000 times transactions onchain, ZK tech will truly shine.

But the overarching question is whether this will occur through the current collective of ZK rollups, or whether ZK tech will be implemented into existing L2s or the base layer. The latter outcome looks increasingly likely. With growing concerns about Ethereum’s positioning and the problem of L2 interoperability, ZK integrations look set to be the key enhancement to improving the execution of the L1. Or, as the key standard that all Ethereum L2s should agree to, as explored in Justin Drake’s ‘Beam Chain’ proposal to introduce ZK execution to the base layer, or Martin Koeppelman’s proposal for 128 Ethereum-approved ZK rollups that all share the same standards.

ZKs will eventually live up to the hype, but it doesn’t look like the first movers will benefit.

Odds & Ends

• Safe launches new L2, Safenet for cross-chain execution Link

• Flashbots and Beaverbuild launch BuilderNet Link

• Sky Ecosystem Risk & Analytics Dashboard Link

• Decentralised.co launches SnetientMarketCap as dashboard for AI agents Link

• Implications of Tornado Cash’s victory in court Link

• Hyperliquid airdrop praised for wide distribution and strong performance Link

• DeFi TVL hits an all-time high over $200bn Link

• Ethena continues USDe growth with new partnership Link

Thoughts & Prognostications

• What I would love to see in a wallet [Vitalik]

• Ethereum’s North Star [Jon Charbonneau/DBA]

• The return of the basis trade [Charlie/Parsec]

• Understanding staking yields on Ethereum and Solana [Tanay Ved & Matías Andrade/CoinMetrics]

• Memes tokens and consent [Dan Finlay/Metamask]

• The Opinionated Application: The App-Infra Thesis, an ode to customisation [Rain&Coffee/Maven11]

• The 5 levels of secure hardware [Georgios Konstantopoulos/Paradigm]

That’s it! Feedback appreciated. Just hit reply. Brr! It’s cold in Tennessee. Enjoyed work and play in Asia and seeing readers at Devcon.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. *I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO/Sky. Some of my compensation comes from MKR, so I’m financially incentivized for its success. All content is for informational purposes and is not intended as investment advice.

DeFi began as a meme during the 2019 bear market and quickly rose to prominence. And with the explosive DeFi Summer in 2020, it soon became the leading crypto narrative outside of Bitcoin. At first, DeFi was hailed as the future of finance, the ultimate 'killer app' for crypto. However, as the bull market surged in 2021, it was NFTs (with their playful JPEGs) that captured mainstream attention. 

Since then, DeFi has struggled to regain momentum, awaiting a breakthrough product that has yet to materialize. Few standout DeFi products have launched post-2020, and the OG DeFi tokens have failed to keep up with L1 tokens (only Aave has held its own). No consumer-facing application has achieved widespread adoption, and institutional DeFi has also failed to take off. 

Despite the stagnation, DeFi activity continues its slow ascent, and is still the most widely used application type on blockchains. Moreover, unlike centralized players, the largest DeFi protocols have never failed. So why have investors and builders soured on DeFi? 

Are we in the early stages of the decline of DeFi? 

Our short answer is no. DeFi isn't disappearing— but it seems to be diverging. One side is more focused on attracting regulated U.S.-based investors, while the other embraces a borderless, permissionless ethos for global users. This bifurcation is making product development more focused and mission-driven for both segments.

Below, we explore the factors behind DeFi’s rise and then slowdown, and discuss whether stablecoins, RWAs, and other centralized assets will kickstart a DeFi Renaissance. 

The core innovations of DeFi

DeFi distinguishes itself from TradFi with four key properties:

1. Non-custodial

2. Composability and permissionless

3. Transparent collateral and trustless execution

4. Onchain liquidity pools.

Though it may just be heresy in Ethereum-land, Bitcoin was the first official DeFi product, because it offered the first way to save without relying on an intermediary. Ethereum took this a step further by enabling non-custodial financial services built using smart contracts. Even now, this remains the defining feature of DeFi. This doesn’t mean that everyone must “be their own bank”, but it does mean that in DeFi, custody of assets is decoupled from financial activity.

“Money legos” were one of the first mini-memes in DeFi that demonstrated how different DeFi protocols could be packaged together permissionless-ly to form a new product. As DeFi has gone multichain, composability has become harder, although progress has been made at the infrastructure layer over the past year.

DeFi has shined most during periods of market turbulence, where anyone can inspect the onchain collateral of DeFi lending platforms, in sharp contrast to the black box of centralized lenders. This was most evident when Celsisus paid-off their onchain debts first; you can’t sweet talk a smart contract to hold off for a few more hours. Likewise, something like FTX repurposing (ahem, stealing) customer funds literally couldn’t happen in DeFi. This is because balance sheets in DeFi are transparent and its rules are enforced by open-source code. 

The last major innovative property of DeFi are pooled onchain assets that enable liquidity for small markets. The first wave of DeFi lenders and decentralized exchanges struggled to gain users because of the cold-start problem. In lending, Compound pioneered liquidity pools so users didn’t need to match with a specific lender. Rather, they can now simply deposit into a liquidity pool where deposit and lending rates are determined algorithmically. For DEXs, Bancor and later Uniswap introduced automated market makers that utilized onchain liquidity pools to facilitate trading. Blue chip assets may be more efficiently traded on orderbooks, but onchain liquidity pools allow the long-tail of assets to easily bootstrap liquidity with low amounts of capital committed, and without needing to hire a market maker. Liquidity pools are key infrastructure for oracles and onchain liquidity is crucial to decisions around listing tokens on lending platforms. Liquidity pools also unlock another key innovation in DeFi: flash loans.

What went wrong?

With such innovations, why hasn’t DeFi already overtaken TradFi? 

In part, it’s still too early. Much like how driverless cars are now slowly coming to market more than ten years after the hype began, building confidence and mainstreaming financial markets takes time. DeFi is not social media, where breakaway growth comes right after the proof of concept. Getting people to invest requires a higher level of trust that will take decades to build.

Next comes the elephant in the room. The regulatory environment in the U.S. – which has been broadly hostile towards crypto over the past three and half years – has dented DeFi in three notable ways:

1. Lack of quality assets: A financial system is only as good as its assets, and there aren’t enough blue chip assets to bring in more investors. DeFi can never reach real-world impact if it's constrained by the total market cap of crypto assets.

2. Banned front-ends for U.S. users: Many DeFi applications - like (Sky, formerly MakerDAO*) have geoblocked U.S. users due to fears of regulatory appraisal. Importantly, this prevents new U.S. companies from building mainstream-friendly products that use DeFi in the backend.

3. Disengaged DeFi governance: There’s no sadder story than Compound, which ushered in onchain money markets and kick-started DeFi summer with the launch of COMP. Its core team has moved on, and while they do not state it directly, fear of SEC action has kept their hands tied. This means that DeFi protocols suffered from a vacuum of governance with no serious engagement from investors and stakeholders.

This has hindered the growth of successful DeFi projects and prevented adjacent centralized players (CeFi, TradFi and traditional consumer tech companies) from offering products that would appeal to the crypto naive. 

There are two other main impediments to DeFi’s growth:

1. The pernicious effects of MEV, with MEV-fear having stopped more crypto natives from moving onchain en masse.

2.  The difficulty in building healthy credit markets. 

DeFi lending is almost entirely dependent on leverage for underlying demand. It’s simply not a big enough market, and the attempts to expand to new areas like uncollateralized lending have largely failed. 

So to recap, DeFi has failed to attract new crypto users largely because of regulatory constraints and a lack of quality assets. And it has failed to grow organically because of MEV and the inability to substantially grow credit markets.

DeFi growth: Two divergent paths

While it seems the regulatory winds will change after the U.S. election regardless of the outcome — and this will help with overall enthusiasm around onchain products — this won't help DeFi attract more high-quality assets. Even if there is a warmer regulatory reception to the idea of decentralization, there is no appetite for allowing tokenized versions of stocks or other real world assets to be freely traded by U.S. persons (unless that activity is under the regime of U.S. security laws). The same is true for any product offering a yield.

With U.S. regulation inevitable, there seems to be two future versions of DeFi emerging. One for U.S. investors that is regulated, and one for the rest of the world, which is less so.

The foundations of this regulated future are now emerging, with the issuance of tokenized regulated funds that limit purchases to qualified individuals and institutions. Here, the tokens come with a centrally-controlled “allowed list”, preventing transfers to unqualified individuals. We are already seeing product releases that would fit this model from TradFi and DeFi teams. Earlier this year, BlackRock launched BUIDL, a tokenized treasury product, and Superstate – founded by Robert Leshner, former CEO & founder of Compound – tokenized two regulated funds to be traded onchain. RWA.xyz tracks all of these tokens.

The onchain activity for these tokens is currently very limited. There are only a few addresses holding tokens. There are no lending pools, yield aggregators, or any onchain trading at all, but the benefit of tokenization is the ability to create a market structure that is digitally and globally native. Once there are enough assets onchain, then DeFi protocols are likely to launch, limiting users to regulated entities.

But is this really DeFi? Well, if we look back at the four innovations that DeFi unlocks above, we might give this version of DeFi a B grade. It’s not permissionless, but it can still offer transparency and trustless execution while enhancing efficiency with pooled assets. Perhaps the centralization of asset issuance will simply become the price of admission for U.S. institutional investors.

We can also see this drift towards regulation at the infrastructure level. ZKsync is actively marketing itself as an L2 for “institutional-grade tokenization” that would help maintain “regulatory compliance”. The launch of Unichain gives Uniswap more regulatory flexibility by controlling the validator set through UNI staking. Uniswap, in particular, seems well-positioned for a more regulated DeFi world with a wallet, an L2, a DEX aggregator, and a DEX. 

The tokenization craze…again?

All of this sounds a lot like the tokenize-everything wave that swept the crypto space in 2017 (remember tokenized parking spaces and security tokens?). Much of this was silly, but the wave really fell short because there was never any reason to use a blockchain to facilitate regulated activity. Now with DeFi, we can see the attraction of financial infrastructure unique to blockchains. Onchain protocols can be valuable even if they're highly regulated.

There could also be some new unlocks from a more regulated DeFi. A regulated future reopens the possibility of capital formation returning as a key value add. If tokens had clearer guidance on how to be traded, then we could see the return of initial coin offerings (ICOs), which paired with AMMs and lending protocols could really lower the cost of liquidity and financing to small organizations. DeFi can compete for non-crypto businesses if there is a regulated, cheaper path to liquidity through onchain markets.

Two DeFi worlds, one DeFi Renaissance

At this point, it seems inevitable that there will be two versions of DeFi. So how will they interact? 

Will they be at odds?

Ever the optimists, we say no. In fact, these two worlds will be largely complementary. We are already seeing this with stablecoins. The biggest holder of the BUIDL tokenized treasury product from BlackRock is Ondo Finance, which uses it to back another stablecoin (USDY) and pass along the yield from BUIDL’s holding of U.S. treasuries. Even Sky (formerly MakerDAO) has moved away from onchain backing of the USDS stablecoin (an upgraded version of Dai). Its RWA Tokenization Grand Prix invites teams to compete to introduce more RWA into USDS’s collateral base. M^0, meanwhile, is a new project that allows minting and redeeming of its M stablecoin by qualified financial institutions with appropriate backing.

Oddly enough, the most appealing DeFi product for non-U.S. individuals is stablecoins tied to the U.S. dollar. Already, it’s easy for U.S. residents to save in dollars, and it’s the asset of choice for the wealthy in developing countries. Yet there is huge untapped demand for dollar savings accounts across the world that stablecoins could tap into by offering cheaper access to dollar yield. This follows in the tradition of the Eurodollar, which are U.S. dollar-denominated deposits at foreign banks that are not subject to U.S. banking regulations. Eurodollars typically offer higher yields than dollars held in U.S. banks, but are also riskier because they don’t have the same reserve requirements or FDIC insurance.

With Stripe’s acquisition of Bridge, stablecoins are on the precipice for a breakout moment that will fuel growth in both regulated and non-regulated DeFi. This will bring in two new DeFi users: businesses that get stablecoin-pilled from Stripe, and retail savers outside the U.S. who don’t want to speculate but rather save using the world’s most stable asset. These users will spur new DeFi products not built around leverage nor speculation.

We expect the regulated version of DeFi to grow faster because it taps into the oceans of capital at regulated institutions, but the non-regulated DeFi world will still thrive, as has been clear from the success of Binance and Tether. Not only will this world serve a different market segment, like the Eurodollar market it will come with higher yields and more risk. The greater risk appetite in this world will also drive product innovation that will push innovation in regulated DeFi forward. This happened with Ethena, a synthetic dollar that derives yield from the basis trade, which surged to a $3 billion market cap in five months despite being off limits to U.S. investors. A month later, Superstate launched a regulated version of this with the Crypto Carry Fund. Non-regulated DeFi will be able to move quicker, but regulated DeFi will have access to an ocean of capital.

So how will it play out? Predictions are fun, so we’ll make a couple. In five years, the TVL of regulated DeFi products will be 50% of all financial activity on blockchains. And in ten years, the market will stabilize with regulated DeFi at 70%, as more and more real world assets are tokenized and the ocean of capital from regulated U.S. investors pours in.

The question is, will we still call it DeFi? TBD, but we believe two things will be true. 1. Financial activity will be the primary value driver for blockchains 2. Blockchain-based finance will eventually surpass traditional finance in scale.

Call it whatever you want.

Odds & Ends

• African crypto startup Yellow Card raises $33 million in new funding Link

• Overview of recent stablecoin bills Link

• Uniswap launches its own DeFi-focused L2 Link

• Recent trends in DEX volume across multiple chains Link

• Flashbots releases Rollup boost that utilizes TEEs Link

• Maker waffles on rebrand to Sky Link

• Kraken plans to launch its own L2, Ink, based on Optimism Superchain Link

• SAFE announces multichain gasless deployment for same address Link

Thoughts & Prognostications

• Possible futures of the Ethereum protocol, part 3: The Scourge [Vitalik]

• A new era of DeFi with Application Specific Sequencing (ASS) [Yuki Yuminaga/Sorella Labs]

• Money routers- the Economic case for bridges [Joel John/Decentralised.co]

• 2022 Seed Stage Retrospective [Lattice]

• How reliable is your node? [Kerman Kohli]

• Polymarket and the power of collective intelligence [Matías Andrade Cabieses & Tanay Ved/CoinMetrics]

• Uniswap v4 and Onchain Investment Banking [Peteris Erins/Auditless Research]

  ---

  That’s it! Feedback appreciated. Just hit reply. Fall is here! I’ll be in Bangkok for Devcon and the week before. If you want to meet up, get in touch!

  Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. *I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO/Sky. Some of my compensation comes from MKR, so I’m financially incentivized for its success. All content is for informational purposes and is not intended as investment advice.

While the weather is starting to cool, the discussion around Ethereum’s future has heated up. Solana’s breakout year and concerns over extractive L2s have shaken confidence in Ethereum. While prices and the market remain stagnated, progress has come on the research side. Below, Zhev looks at the MEV space from the perspective of order-flow auctions (OFAs), and evaluates what could well be the final MEV boss: running credible auctions on censorship-resistant blockchains. 

Collectively, we’ve progressed rapidly towards the MEV-mitigating goals set out by Flashbots and co-opted by Ethereum. In fact, the success of Flashbots – practically the face of MEV on Ethereum – implies that the illumination of the dark forest goal has almost been met. The other two – democratization of mindful extraction, and the distribution of benefits – are also seeing more daylight. However, there remains more work to be done before Ethereum reaches the MEV utopia, where leakage is minimized at every stack of the protocol. 

The easiest gotcha for MEV on Ethereum is that most auctions are entirely offchain and centralized. This means that the protocol cannot give any guarantees regarding execution to any participant in such auctions. Rather, guarantees are bestowed by a second agent who is more privileged in the mechanism. 

This is most clearly illustrated with Order Flow Auctions (OFAs). Such auctions were supposed to be a solution to the problem of value distribution, a way for users to recapture the value they’re responsible for creating by running an auction antiparallel to MEV-Boost, or any such similar PBS auctions. And while OFAs have enhanced user welfare, there are other issues. OFAs rely on trusted intermediaries who are ultimately at the behest of the oligopolistic builder market, which has proven willing to censor transactions as they please. As such, efforts must be made to ensure that block producers aren't able to influence applications’ transactions and alter them in their own favor. 

This ultimately means that new OFA designs that are decentralized are needed. And more importantly, so is credible infrastructure with censorship resistance to run these OFAs, be it SUAVE and FOCIL or the newly introduced BRAID – which would introduce multiple proposers to the Ethereum protocol. Although SUAVE and a Flashbots/Ethereum-aligned future seemed like the inevitable MEV endgame, the route that brings multiple proposers to Ethereum looks like the unexpected favorite. 

Order flow auctions, but make them credible

In a previous article, we evaluated some of the emergent OFA platforms at the time, which serve as trusted intermediaries between extractors (searchers and builders) and users under the PBS framework.

Before we move on, it’s important to note that most performant OFAs to date – such as UniswapX, CoWswap, and the like – are application-specific (in this case, they offer OFAs for trade/swap execution). This means they are not generalizable MEV infrastructure but specifically designed to prevent frontrunning retail traders, so while valuable, it’s not a long-term building block that offers the programmability of say a smart contract. This also doesn’t even factor in other pain points such as the cost of bootstrapping a solver network that actually prioritizes user welfare (promise you won’t front run anyone, bro), and the disadvantages that come with a siloed solver network.

There’s also the problem that most OFA platforms thrive on a payment for order flow (PFOF) or exclusive order flow (EOF) model. Here, user-facing applications such as RPCs, wallet providers, and decentralized applications receive payments from extractors for exclusive access to the users’ orders, a sort of vertical integration.

So then, we have two key problems:

1. the proliferation of backroom deals and a PFOF model that isn't optimal for user welfare.

2. block producers (builders and proposers) can censor transactions which pay them only a minimal portion of the transaction's extractable value. In other words, the base protocol's low cost of censorship.

To overcome the first problem, application-agnostic/generalized OFAs are being designed with the goal of being implemented as a portion of the protocol's execution logic (just like MEV auctions), so that applications can easily implement OFAs with lower overheads. This would ultimately redefine how user welfare overlaps with inclusion/execution guarantees.

The overarching disadvantage of conducting offchain auctions is that the disincentive for predatory behavior isn't native to the system, but rather enforced through loss to reputation and/or fixed penalties applied by a (centralized) authority.

Also, besides other factors (such as participation costs) which invariably lead to an undesirable oligopolistic extractor regime in MEV markets, the PFOF model is unavoidably oligopolistic, as demonstrated by the increasing oligopoly of market making in TradFi by HFT firms.

This oligopoly isn't achieved through fair competition among extractors, where they optimise for features and user welfare. Rather, via backroom deals which the user doesn't necessarily benefit from, and might not even be aware of. In fact, 80% of block builder revenue comes from private order flow - so why take away the punch?

In order to stop this feedback loop, work must be done to introduce more competition and decentralization in both the OFA market and in the block builder market.  

The most important thing to fix is the centralized builder market, which prevents MEV auctions from being run onchain. We’ll discuss that in more detail below, but wanted to highlight two promising OFA designs, MEV-Share and Atlas, which make improvements on two key areas: privacy and application sovereignty.

• Programmable privacy via MEV-Share:

  • MEV-Share is implemented as an open source pipeline for execution/inclusion OFAs striving for mindful extraction. 

  • Originators send transactions to the MEV-Share node, which allows them to specify privacy preferences using hints. These hints are transaction-specific data parameters including call data, logs, hash, contract address, etc. The node is responsible for stripping the user's order based on the hints they choose to reveal, producing a hash of this stripped/altered order, and sharing the hash with integrated searchers as events. From here on, it functions like a typical OFA, where searchers compete for execution rights, then inclusion rights in an OFA bundle, before proceeding to the PBS auction for payload execution.  

  • MEV-Share’s key distinction is its privacy guarantees while maintaining a permissionless auction entry, which could be thought of as a new public mempool. This hints at the larger vision with SUAVE and TEEs. 

• Application-specific sequencing via Atlas:

  • Atlas is a generalized execution abstraction protocol for application-level MEV-minimization, which coalesces solver networks across applications and allows them to host all sorts of specialized OFAs.

  • Atlas enables what is referred to as application specific sequencing, which just means that it allows applications to exert granular control over their order flow. This allows the crafting of more bespoke designs that better reflect the needs of an application and its users.

  • It is based on the concept of “execution abstraction”, and leverages EIP-4337, which was created to give smart contracts greater functionality. Atlas leverages this secondary mempool to bundle 'value creating' operations inside of the same transaction as the 'value capturing' operations, so nobody can 'unbundle' it or censor any of the bids without also censoring the piece that creates the value. 

Share vs. Atlas: Infrastructure lacks either way

The primary design difference between MEV-Share and Atlas lies in where the auction's inclusion guarantees flow from. The former relies on guarantees from block builders to be credible, so decentralization and competitiveness of the builder set is very important, which Flashbots says is why they’re building SUAVE. 

Meanwhile, Atlas chooses to use an Entrypoint contract, akin to ERC 4337, to get direct access to an alternate mempool (called ‘ops relay’) hosted by an interested party. This difference in the source of the inclusion guarantee allows Atlas to be much more flexible, as it doesn't need a PBS-supportive protocol to be performant. However, this comes at the cost of its gas efficiency since there's a lot of verification to be done onchain.

These designs improve user welfare, but ultimately, they will only work if there is credibly neutral infrastructure to run these auctions.

OFAs and the censorship-resilience inevitability

In “credible, optimal auctions via blockchains” Tarun et al. examine the effects of blockchains and cryptography on credible auctions. The notion of credibility in this context is based on Akbapour's work where a credible auction is defined as one in which the auctioneer's (seller's) revenue is better off when they follow the described mechanism. The results from Tarun et al. show that censorship-resistant blockchains fuel credibility.

The question is can a blockchain run a fair, high-value auction while maintaining censorship resistance? In the current setup, Titan, rsync and beaverbuild (the three largest block builders on Ethereum) have a privileged position for any onchain auction and will simply censor any transaction from an auction that tries to redistribute MEV they have dutifully earned. This is a huge obstacle to efforts to reduce LVR and the CEX-DEX arbitrage. Thus, any credible OFA would only become fully performant when there are guarantees of fair inclusion and censorship resistance against all odds from the blockchain. 

The prevalent questions, then, are (a) how can Ethereum (or another blockchain) provide these guarantees? and (b) in what ways can the power of a block producer be limited to ensure there is no ability to censor? 

There are two paths emerging for where the market will go. The first is a continuation of the MEV mitigation and redistribution strategy of the last four years, which has been led by Flashbots. And the second path tries to fix the source of MEV privilege by removing the proposer monopoly entirely. 

SUAVE, FOCIL, Flashbots and the Ethereum Foundation

Flashbots launched with a charge to “frontrun the MEV crisis”. This was a tacit acknowledgement that they weren’t going to fix the crisis, but rather build extractive tools and then figure out how to make them fair.

SUAVE (or the Single Unifying Auction for Value Expression) is the endgame for this vision, a blockchain that is a decentralized block builder for any EVM chain.

SUAVE extends MEV-Share's programmable privacy to execution environments which can use it as a decentralized block builder, or even a shared sequencer. It utilizes “TEE-kettles”as confidential compute enclaves, and runs as “SUAVE chain” powered by Clique's proof-of-authority consensus protocol. SUAVE intends to be the credible infrastructure for MEV auctions, essentially becoming the home of all MEV extraction, but with the upside that it’s being run on a decentralized blockchain, rather than the opaque MEV supply chain of today.

This approach offloads the problem of censorship resistance to a different environment than the L1, which means Ethereum’s censorship-resistance would be dependent on another chain.

To shore up its censorship-resistance, Ethereum researchers have proposed inclusion lists. While the specifications differ between designs, the basic premise of inclusion lists is to allow the proposer to forcefully include some transactions in their slot (or a future one!), potentially against the wishes of a censoring builder. 

FOCIL takes this a step further to move the production of an inclusion list from the slot's proposer, to a leaderless committee of validators selected randomly. For every slot, a random set of validators produce a local inclusion list from transactions in the mempool, which must be included in the next block.

This design is more of a patch over one of the many leaks of PBS on Ethereum, and the thing with patches is that they mostly don't last. The team at init4 tech also recently showed that forced inclusion doesn’t prevent censorship of most DeFi transactions. FOCIL may be useful but it will not address the builder monopoly.

So path one is where Ethereum implements something like FOCIL at the protocol level and then relies on SUAVE to decentralize its block building market.

BRAID, the darkhorse

Flashbots and the Ethereum Foundation have both agreed to work to externalize MEV auctions outside of Ethereum, but that conventional wisdom is beginning to fray with the emergence of multiple-concurrent proposer (MCP) designs and BRAID.

With multiple proposers, rather than having a single proposer append blocks for every slot, the protocol implements a leaderless scheme in which at least two proposers are responsible for producing the payload to be executed for the slot. This removes the monopoly a single entity has on inclusion and allows the protocol to expose a more expensive cost of censorship at every slot, more than it would have in a leader-based scheme. This is due to the observation that a single slot with K proposers achieves the same cost of censorship that would require K slots on a single proposer chain. 

The work of Max Resnick on BRAID has again sparked interest in the topic as a viable means for increasing the censorship resistance/cost of censorship of a protocol. While the specifications are still being fleshed out in real time, BRAID and its multiple-proposer architecture is a wrecking ball to the current Ethereum roadmap and its PBS architecture (it’s seen as a direct alternative to FOCIL). After being presented last month, BRAID quickly gained support from Dan Robinson of Paradigm, which is notable given their investment in Flashbots, and SUAVE by extension. Not everyone from Paradigm is on board however, and it looks like SUAVE will compete with BRAID for what will ultimately be the long-term solution for MEV mitigation. Max was even on Bankless twice in six weeks trying to shift the Ethereum mindshare (and it appears to be working – at least on the “Are L2s extracting?” discussion).

There’s also been other criticism of BRAID, which is aimed at its idea of a leaderless scheme for consensus. In such a system, there has to be some allowance for latency, so that proposers can achieve some extent of simultaneous release. A short duration would lead to missed slots and potential liveness failures. But longer ones would expose the last-look problem, where agents can delay in order to view the blocks released by other proposers and potentially grief them.

If implemented, BRAID would upend the MEV supply chain. There would still be leakage but no longer a clear actor in the system who could exploit it. This would mean more redistribution based on competitive dynamics rather than goodwill.

Is there use in trying?

MEV has proven such a vexing (and interesting!) problem to solve, we often forget why it’s so damaging to user welfare in the first place. Users now fear transacting onchain because of the predatory behavior of MEV. Indeed, if you have to trust someone, why not opt for CEXes? They’re more friendly than the creatures of the dark forest. That sentiment will not change overnight, but the work being done now gives application developers the tools and infrastructure to make MEV invisible without new centralization risks.

This would be a huge step forward, but would it ultimately fix the problem of private order flow?

Probably not. But it will fix exclusive order flow where entrenched builders purchase flow and further cement their builder monopoly. Instead, without a centralized builder market, the problem of private order flow becomes a question of best execution for application developers. 

With credible infrastructure to run auctions that cannot be censored by block builders, applications and other transaction originators will be more in control of their MEV supply chain. This means they’ll have to become more adept at bundling transactions or outsourcing to third party builders who do. The role of tightly packing blocks will still be needed but done by distinct actors at different parts of the supply chain.

Odds & ends

• USDS and SKY launch on Ethereum, completing MakerDAO rebrand Link

• Centrifuge launches new institutional RWA Morpho market on Base Link

• Visa releases new dashboard on stablecoin usage Link

• House lawmakers clash at DeFi’s first congressional hearing Link

• Coinbase’s cbBTC reaches $120m one week after launch Link

• SAFE proposes native swap fee Link

• Doppler, a liquidity-bootstrapping hook design on top of Uniswap v4 Link

• Atlas, a high-performant DeFi Ethereum L2 based on SVM, launches testnet Link

• ZKsync announces onchain governance system Link

Thoughts & prognostications

• How to win friends and TEE-fluence people [Ethan Buchman]

• The week in crypto policy: Fall preview [George Leonardo/Messari]

• Abstracting chains [Shlok Khemani/Decentralised.co]

• State of stablecoins: Sector expansion & a changing interest-rate environment [Tanay Ved/Coinmetrics]

• Ethereum has gotten stronger [Vitalik]

• From fragmentation to cohesion: Coordinating last miles to unify transaction supply chains [Kevin Wang/Khalani]

• The block builder landscape [Facundo Indabera/Delphi Digital]

• Observations on factors driving low gas prices [Teck Yuan Lee/Etherscan]

That’s it! Feedback appreciated. Just hit reply. Great to be back after a restful summer!

Dose of DeFi is written by Chris Powers, with help from Denis Suslov, Zhev and Financial Content Lab.

Now, the mindshare of researchers, builders and investors turns to solving the thorny problems that prevent mainstream adoption. Unresolved MEV issues and the lack of regularity clarity are hurdles to overcome, but the primary obstacle to mainstream adoption is the poor UX from a lack of chain interoperability, especially among Ethereum L2s. To achieve frictionless interoperability, researchers and builders have coalesced behind the abstraction meme, be it account or chain abstraction. Below, Zhev looks at how these aim to improve UX, and which meme will ultimately win out.

- Chris

Blockchain builders are stubborn, with strong opinions. This has led to the dominance of ideologically-driven development that is rigid, and as such, esoteric debates loom larger than the user preferences of the market.

Divergence among these ideologies has ostensibly led to the fragmentation of capital and innovation across siloed ecosystems. This certainly makes things spicy and occasionally grants us front-row seats to spats of drama between these ecosystems. Yet there is no doubt that the future of the “new internet” will be far less tribalistic than it currently is. Eventually.

The next billion users won't be concerned with underlying ideologies. Rather, they’ll care about the outcomes of their transactions and ultimately UX. Yet the current fragmented landscape hinders developers trying to build seamless experiences.

Interoperability is key to enhancing UX, but bridges are too clunky in their current state. A broad reworking of mechanisms at the very base layer – where most of the problems emanate – will be needed for an interoperable future to remain feasible. 

In this piece, we evaluate the basic properties of blockchains that cause fragmentation, and provide an overview of prominent mechanisms being explored to solve this (within and beyond single ecosystems). In this way, we take a bottom-up approach to understanding the latest meme craze of abstraction at both the account and chain levels. And how these influence the interoperability discussion.   

The basics of blockchain state models

Blockchains are notoriously referred to as a shared ledger, but those in the know call this ledger the blockchain state. On blockchains, a state is not static. It’s always being updated with additional transactions. 

How these transactions are added to the state (or ledger) is called a state model, which decides what is a valid transaction, how consensus is reached, and how to maintain a shared ledger state across an honest majority of a blockchain's nodes. Most importantly, the state model specifies how a chain should handle conflicting state updates. For example, when two transactions want to spend the same asset, or when two conflicting transactions want to access the same state.

Blockchain interoperability means a single user initiates a transaction that affects the state model on two separate chains in some trustless and decentralized way. Not all state models are the same. In fact, the two largest blockchains have vastly different structures.

The two major state models are:

1. Unspent transaction output (UTXO) model: State is managed by tracking individual transaction outputs (represented as assets) – that can either be spent or not – and the restrictions on spending them (represented as the asset's holder). Spent outputs and their corresponding inputs are trackable to a public key, which can be used to approve said input for future spends by verifying key ownership. Thus, each asset is tracked using the last transaction in which they were spent, and the current restrictions on spending them again (its current holder).

2. Account model: Maps an address to its balance and current state. As opposed to tracking each asset in the network, this model tracks each account in the network and balances for each asset on a unified ledger. All interactions in an account-based chain are implicitly interlinked, so there’s only one global tree of interdependent transactions which are both state and time-variant.

Bitcoin uses the UTXO model, with limited programmability. However, this drawback comes with various execution advantages. The most notable is transaction localisation, which means that instead of creating a new branch as part of a global state tree for each interaction (as in the account model), isolated state transitions can be represented and executed as an isolated local state tree.

The account model, meanwhile, is mostly used by Ethereum and other smart-contract chains. It’s less rigorous because it leaves determination of a transaction's state to smart contracts. This makes things a bit complicated as smart contracts mostly specify state changes based on the chain's current state. In the presence of multiple contracts, this can easily lead to unexpected outcomes if the current state is modified by another transaction before the smart contract's desired state change can be attained.

The key takeaway: as long as settlement occurs on a global state tree with a single-threaded and sequential execution scheme, there will always be implicit interdependence between smart contracts. This means that transactions are neither time- nor state-invariant, and can be invalidated due to another valid transaction that comes before them. It also means a “double spend” isn't apparent until a contract is executed – as opposed to a UTXO model, where an invalid state change could be determined without using any of the execution layer.

Accounts on Ethereum

The widespread adoption of the account model makes it an ideal component to build interoperability infrastructure. Yet to do this, a new perspective around accounts, rather than blockchains themselves, is needed. This certainly seems consistent with the broader trend: Stephane Gosselin of Frontier Research recently compared the shift from a chain-centric view of crypto to an account-centric view to the shift from a geocentric worldview to a heliocentric worldview.

Ethereum accounts are decomposed into four fields: a nonce, a balance, a codeHash, and a strorageHash. These help define the two types of accounts in Ethereum: externally-owned accounts (EOAs) and contract accounts (CAs). EOAs can be controlled by anyone holding the private keys and have empty codeHash and storageHash fields (and by extension, don't have access to persistent storage). 

CAs are controlled by code deployed on to the network (aka, a smart contract). Their interactions are entirely pull-based, i.e., they can only send specific transactions in response to a transaction from another account. Thus, the actions of a CA are somewhat static and predetermined by its code's logic. They have to be built/derived from an account creator's address and its nonce value, and the account creator must be an EOA account with a balance to pay for the account's storage usage. 

In short, EOAs provide users with full autonomy (i.e., they can begin transactions) at the cost of programmability (i.e. their logic is strictly constricted), while contract accounts take the other route (programmability at the cost of autonomy).

Transaction validity

Achieving interoperability is important because to operate on a new chain, a third party must send a transaction on the user’s behalf. Of course, blockchains are built to prevent this, so new protocol rules will have to be introduced. 

Currently, for transactions from any account to be considered valid and then executed, there are four checks (or what we call “constraints”) it must satisfy: 1) Authentication 2) Authorization 3) Replay protection, and 4) Gas payment.

For EOAs, each of these checks are constrained to ensure a single valid result. Authentication and authorisation are provided by the EOA’s private key; replay protection is provided by the transaction's nonce, and gas payment must be provided by the account from which the transaction originates. These enshrined constraints limit EOAs. 

Contract accounts allow more programmability around these validity checks. Authentication and authorisation of transactions are dependent on its code's logic, granting a lot of leeway around key management and access policies (gas payment may be settled by another account, e.g., its caller). The lack of granular constrictions allows transactions originating from contract accounts to be more programmable, granting them coveted features such as atomicity.

Building an account layer

Okay, now that we understand the general structure of accounts on Ethereum and what is needed to send a valid transaction, we can look at the account-based technical solutions aiming to enhance interoperability. 

Like Bitcoin, Ethereum launched only with EOAs but first imagined “smart accounts” in 2017 with the introduction of EIP-86. Smart accounts are supposed to combine the best properties of EOAs and CAs in order to birth a new, composable primitive, but the process has been slow and tedious.

The challenge for the Ethereum community is needing to upgrade two different types of accounts (EOAs & CAs). Although researchers said they were moving forward with both, it always seemed like upgrading EOAs was less of a priority, given the view of many - including Vitalik - that smart contract wallets are the endgame for regular users. 

A vision for this smart contract wallet-led future began to coalesce around the idea of Account Abstraction, which would make all accounts on Ethereum contract-based. AA would remove all deterministic constraints on accounts from the consensus layer, and define probabilistic constraints within the execution layer, which would create a single standard for accounts across the EVM world.  

The cornerstone of this vision is ERC-4337, which was implemented last year and allows accounts to package up the actions they want to take into a UserOperation, which is sent to a dedicated mempool where bundlers can put them all into a transaction that makes a single call.

A dedicated mempool is a huge step but achieving the full vision of AA would require a lot of changes to the Ethereum protocol, including the consensus layer, which is always a sensitive area. As such, Ethereum has been slow in implementing upgrades to its account model that would enhance interoperability.

Balancing future desire with current needs

So even if smart contracts and AA are the future, EOAs are still in wide practice today so many application developers have been pushing for upgrades alongside the push for protocol changes to enhance smart contract wallets. 

ETH researchers listened and planned EIP-3074 for inclusion in the next Pectra update, but received significant pushback from supporters of EIP-4337. They argued the new transaction type proposed would be incompatible with EIP-4337 and solidify the two parallel track models for EOAs and CAs. There were also concerns about increasing censorship but the important point was that EIP-3074 would hurt the long term vision of Account Abstraction and EIP-4337. There also may have been competitive worries about getting new users on smart contract wallets if EOAs had the same functionality.

Responding to the pushback, EIP-3074 was withdrawn and Vitalik wrote EIP 7702 in 22 minutes to achieve the same ends as EIP-3074 but is forward-compatible with EIP-4337, and importantly, does not introduce any new opcodes to the EVM.

Account Chain Constraint abstraction 

 We believe there is a middle ground that maintains Ethereum’s gradual approach to achieving AA yet also enhances the experience of EOAs, so that interoperability infrastructure can be built now. Given that upgrading EOAs with more programmability means that only some of the deterministic constraints are removed from existing account standards, we refer to this as “constraint abstraction”.

Constraint abstraction is considered a viable alternative due to many factors: including, the complexity associated with absolute abstraction, EOA enshrinement in dAPPs’ UI/UX, and users’ fears.

Chain abstraction

Having parts of an account as modules rather than irreversibly integrated verticals opens up various channels for interoperability. These channels that can transcend chain boundaries represent a new approach to solving the fragmentation problem, and have been collectively referred to as chain abstraction.

Similar to account abstraction – an intended end state with numerous pathways to its realization – chain abstraction seeks to unify and simplify users’ actions across all (most, if you seek to be pragmatic) existing ecosystems. While it’s easy for any one chain to point fingers and proclaim that their stack is better, it should be all too obvious at this point that the best way forward is a collaborative effort for UX simplification.

OneBalance, which emerged out of Frontier Research, was created with this purpose: a collaborative that transcends execution environments to provide a single account standard for users across all ecosystems.

It describes its proposed solution – credible accounts – as “ultra HTLCs”, and we tend to agree. HTLCs are a great primitive that never got their medals due to the complexity associated with implementations when first ideated. Maybe now we have a reliable mechanism to abstract this complexity from users.

Intents, again

Without clear technical definitions, the difference between account, chain and constraint abstraction is mostly on the meme side of things. What is apparent is that whatever direction the industry moves, the new system will be powered by intents and the assumption that a set of more specialized agents – driven by some definable amount of (non-detrimental) extractable value – would be willing to compete to help users settle their requests.

While there are various fanciful definitions for intents, they are simply a user expressing their desired outcome/goal for a broadcasted request. A simplification of their request, removing unnecessary restrictions in the “how” of the request and placing them in the “what”.

Specialized actors referred to as “solvers” then take it upon themselves to identify the most cost-effective pathway (the “how”) that respects the user's goal (the “what”).

Let the solvers deal with the complexity

HTLCs failed to take off when they were first introduced in 2017-2018 due to the perceived complexity. Now more than ever, with the presence of MEV-hardened solvers in the mix, it'd be great to see them get the love they deserve. HTLCs are one of the simplest iterations of a technology powering interoperability; on the other hand the rise and funding of intent-centric designs has shown that user welfare is paramount.

We believe that if crypto wants to take the onboarding problem seriously, then it must prioritize user welfare at every possible level. With interoperability, we believe intent-powered HTLCs are a step in the right direction. We also believe a lot of the current work being branded chain abstraction is unnecessary; most of the solutions have existed in the shadows for a long time due to complexity of use. It is now simply time for synergy.

There has always been a need for a simple UX and a desire to implement interoperability. What’s been missing is the motivation (driven by market demand for cross-chain trading) and the increased capability of solvers to execute the complex cross-chain transactions. Solvers would be a good fit, but there is a risk, abstraction further entrenches block builders.

New innovations must now be encouraged to synchronize with old ones without introducing extra vaporware. And intents + HTLCs seems a great place to start in redefining cross-chain UX.

Odds & Ends

• SEC sues Consensys over MetaMask swaps and staking services Link

• Gnosis to launch threshold-encrypted mempools through Shutterized in July Link

• Balancer begins planning v3 Link

• James Prestwich on the Infinite Jungle podcast from Galaxy Link

• Arbitrum proposes TimeBoost, a new transaction ordering policy Link

• Reading list for Trusted Execution Environments (TEEs) Link

• Curve founder suffers liquidation cascade on CRV-baced loans Link

• Restaked ETH reaches $13bn (3.7m ETH) in market size Link

Thoughts & Prognostications

• Blob Inversion: An investigation into the June 20th 2024 blob spike [Blair Marshall/Block Native]

• Were rollups meant to be on Bitcoin all along? [Peteris Erins/Auditless Research]

• Inclusion Lists Don’t Fix Multi-Block MEV in Attester Proposer Separation [Charlie Noyes/Paradigm]

• Making sense of some niche modular infrastructure [knower’s substack]

• Meme coin mania [Tanay Ved & Matías Andrade]

• The unseen bidding war: Rsync’s strategic maneuvers [EigenPhi]

• The state of Polygon [Saurabh Deshpande & Siddharth/Decentralised.co]

• Dude, where’s my (stablecoin) yield? [Kinji Steimetz/Messari]

That’s it! Feedback appreciated. Just hit reply. Great to have Zhev go deeper at the protocol level. Lots of of great discussion in the Ethereum research community this week as well.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov, Zhev and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO. All content is for informational purposes and is not intended as investment advice.

- Chris  

There’s already been a great unbundling at the core base layer, where Ethereum used to have a single solution for execution, settlement and data availability. However, it has since moved toward a more modular approach, with specialized solutions for each core element of the blockchain. 

This same pattern is playing out in the DeFi lending space. The first successful products were those with everything self-contained. While the original three DeFi lenders - MakerDAO, Aave, and Compound - had many moving parts, they all operated in a pre-defined structure set by their respective core teams. These days, however, growth in DeFi lending has come from a new crop of projects that split up the core functions of a lending protocol. 

These projects are creating isolated markets, minimizing governance, separating risk management, liberalizing oracle responsibilities, and removing other single dependencies. Others are creating easy-to-use bundled products that put multiple DeFi legos together to offer a more all-encompassing lending product.

This new push to unbundle DeFi borrowing has been memed into Modular Lending. We are big fans of memes here at Dose of DeFi, but have also seen new projects (and their investors) try to hype new narratives on the market more for their bags than because there’s some underlying innovation (looking at you, DeFi 2.0).

Our take: the hype is real. DeFi lending will go through a similar metamorphosis as the core base layer – where new modular protocols emerged like Celestia while existing incumbents shifted their roadmaps to become more modular – as Ethereum has done as it continues to unbundle itself.

In the immediate term, the key competitors are forging different paths. Morpho, Euler, Ajna, Credit Guild and others are seeing success as new modular lending projects, while MakerDAO moves toward a less concentrated SubDAO model. Then you have the recently announced Aave v4, which is moving in the modular direction, mirroring Ethereum’s architecture shift. These paths being carved-out now could well determine where the value accrues in the DeFi lending stack over the long term.

Why modular?

There are generally two approaches to building complex systems. One strategy is to focus on the end-user experience and ensure the complexity doesn’t hinder usability. This means controlling the entire stack (as Apple does with its hardware and software integration). 

The other strategy focuses on enabling multiple parties to build individual components of a system. Here, the centralized designers of the complex system focus on core standards that create interoperability, while relying on the market to do the innovation. This is seen with the core internet protocols, which haven’t changed, while the applications and businesses that have built on top of TCP/IP have driven innovation on the internet.

This analogy could also be applied to economies, where a government is seen as the base layer, a la TCP/IP, ensuring interoperability through the rule of law and social cohesion, where economic development then occurs in the private sector built on top of the governance layer. Neither of these approaches work all the time; many companies, protocols, and economies operate somewhere on the spectrum. 

Breaking it all down 

Those that support the modular lending thesis believe that innovation in DeFi will be driven by specialization in each part of the lending stack, rather than focusing on just the end-user experience. 

A key reason for this is the desire to eliminate single dependencies. Lending protocols require close risk monitoring and a small issue can lead to catastrophic loss, so building redundancy is key. Monolithic lending protocols have introduced multiple oracles in case one fails, but modular lending takes this hedging approach and applies it to every layer of the lending stack.

For every DeFi loan, we can identify five key components that are needed – but can be modified:

1. A loan asset 

2. A collateral asset 

3. Oracle

4. Max loan-to-value (LTV) 

5. Interest rate model. 

These components must be closely monitored to ensure a platform’s solvency and prevent bad debt accruing because of rapid price changes (we could also add the liquidation system to the five components above). 

For Aave, Maker, and Compound, token governance makes decisions for all assets and users. Initially, all assets were pooled together and shared the risk of the whole system. But even the monolithic lending protocols have moved quickly into creating isolated markets for each asset, to compartmentalize the risk. 

Understanding the major modular players 

Isolating markets is not the only thing you can do to make your lending protocol more modular. The real innovation is happening in new protocols that are reimagining what’s necessary in a lending stack.

The biggest players in the modular world are Morpho, Euler and Gearbox: 

• Morpho is the clear leader of modular lending at the moment, although it seems recently uncomfortable with the meme, trying to morph into “not modular, not monolithic, but aggregated”. With $1.8 billion in TVL, it’s arguably already in the top tier of the DeFi lending industry as a whole, but its ambitions are to be the largest. Morpho Blue is its primary lending stack, on which it’s permissionless to create a vault tuned to whatever parameters it desires. Governance only enables what can be modified – currently five different components – not what those components should be. That is configured by the vault owner, typically a DeFi risk manager. The other major layer of Morpho is MetaMorpho, an attempt to be the aggregated liquidity layer for passive lenders. This is a particular piece focused on end-user experience. It’s akin to Uniswap having the DEX on Ethereum and also Uniswap X for efficient trade routing.

• Euler launched its v1 in 2022 and generated over $200 million in open interest before a hack drained nearly all protocol funds (although they were later returned). Now, it’s preparing to launch its v2 and reenter a maturing modular lending ecosystem as a major player. Euler v2 has two key components. One, the Euler Vault Kit (EVK), which is a framework for creation of ERC4626 compatible vaults with additional borrowing functionality, enabling them to serve as passive lending pools, and two, the Ethereum Vault Connector (EVC), which is an EVM primitive that primarily enables multi-vault collateralisation, i.e., multiple vaults can use collateral made available by one vault. V2 has a planned Q2/Q3 launch.

• Gearbox provides an opinionated framework that is more user centric, i.e. users can easily set up their positions without too much oversight, regardless of their skill/knowledge level. Its primary innovation is a “credit account” which serves as an inventory of allowable actions and whitelisted assets, denominated in a borrowed asset. It’s basically an isolated lending pool, analogous to Euler's vaults, except that Gearbox's credit accounts hold both user collateral and borrowed funds in one place. Like MetaMorpho, Gearbox demonstrates that a modular world can have a layer that specializes in packaging for the end user.

Unbundled, then rebundled 

Specialization in parts of the lending stack presents an opportunity to build alternative systems that may target a specific niche or bet on a future growth driver. Some leading movers with this approach are listed here:

• Credit Guild intends to approach the already-established pooled lending market with a trust-minimized governance model. Existing incumbents, such as Aave, have very restrictive governance parameters, and more often than not this results in apathy among smaller token holders since their votes seemingly don't change much. Thus, an honest minority in control of most tokens is responsible for most changes. Credit Guild turns this dynamic on its head by introducing an optimistic, vetocracy-based governance framework, which stipulates various quorum thresholds and latencies for varying parameter changes, while integrating a risk-on approach to deal with unforeseen fallouts.

• Starport’s ambitions are a bet on the cross-chain thesis. It has implemented a very basic framework for composing different types of EVM-compatible lending protocols. It’s designed to handle data availability and term enforcement for the protocols built atop it via two core components: 

  1. The Starport contract, which is responsible for loan originating (term definition) and refinancing (term renewal). It stores this data for the protocols built atop the Starport kernel and makes it available when needed. 

  2. The custodian contract, which primarily holds the collateral of borrowers on originating protocols atop Starport, and ensures that debt settlements and closure proceed according to the terms defined by the originating protocols and stored in the Starport contract.

• Ajna boasts a truly permissionless model of oracleless pooled lending with no governance at any level. Pools are set up in unique pairs of quote/collateral assets provided by lenders/borrowers, allowing users to assess demand for either of the assets and allocate their capital accordingly. Ajna's oracle-less design is borne off lenders’ ability to specify the price at which they're willing to lend, by specifying the amount of collateral a borrower should pledge per quote token they hold (or vice versa). It will be especially appealing to the long tail of assets (much like Uniswap v2 does for small-ish tokens).

If you can’t beat ‘em, join ‘em 

The lending space has attracted a slew of new entrants, which has also reinvigorated the largest DeFi protocols to launch new lending products:

• Aave v4, which was announced last month, is awfully similar to Euler v2. It comes after Aave zealot Marc “Chainsaw” Zeller said that Aave v3 would be the end state of Aave because of its modularity. Its soft liquidation mechanism was pioneered by Llammalend (explainer below); its unified liquidity layer is also similar to Euler v2's EVC. While most of the impending upgrades aren't novel, they’re also yet to be widely tested in a highly liquid protocol (which Aave already is). It’s crazy how successful Aave has been at winning market share on EVERY chain. Its moat may be shallow, but it’s wide, and gives Aave a extremely strong tailwind.

• Curve, or more informally Llammalend, is a series of isolated and one-way (non-borrowable collateral) lending markets in which crvUSD (already minted), Curve's native stablecoin, is used as either the collateral or debt asset. This enables it to combine Curve's expertise in AMM design and offer unique opportunities as a lending market. Curve has always driven on the left side of the road in DeFi, but it’s worked out for them. It has as such carved out a significant niche in the DEX market, aside from the Uniswap goliath, and is making everyone question their tokenomics skepticism with the success of the veCRV model. Llamalend appears to be another chapter in the Curve story:

  • Its most interesting feature is its risk management and liquidation logic, which is based on Curve's LLAMMA system that enables ‘soft liquidations’.

  • LLAMMA is implemented as a market making contract that encourages arbitrage between an isolated lending market's assets and external markets. 

  • Just like a concentrated liquidity Automated Market Maker (clAMM eg. Uniswap v3), LLAMMA evenly deposits a borrower's collateral across a range of user-specified prices, referred to as bands, where the offered prices are greatly skewed in relation to the oracle price in order to ensure arbitrage is always incentivised.

  • In this way, the system can automatically sell (soft-liquidate) portions of the collateral asset into crvUSD as the former's price decreases past bands. This decreases the overall loan health, but is decidedly better than outright liquidations, especially considering the explicit support of long-tail assets.

Whew. Curve founder Michael Egorov making over-engineered criticisms obsolete since 2019. 

Both Curve and Aave are hyper focused on the growth of their respective stablecoins. This is a good long-term strategy for fee-extraction (ahem, revenue) capability. Both are following in the footsteps of MakerDAO, which has not given up on DeFi lending, spinning off Spark as an isolated brand that has had a very successful past year even without any native token incentives (yet). But a stablecoin and the crazy ability to print money (credit is a hell of a drug) are just gigantic opportunities long term. Unlike lending, however, stablecoins do require some onchain governance or offchain centralized entities. For Curve and Aave, which have some of the oldest and most active token governance (behind MakerDAO of course), this route makes sense.

The question we can’t answer is what is Compound doing? It was once DeFi royalty, kickstarting DeFi summer and literally establishing the yield farming meme. Obviously, regulatory concerns have constricted its core team and investors from being more active, which is why its market share has dwindled. Still, much like Aave’s wide, shallow moat, Compound still has $1 billion in open loans and a wide governance distribution. Just recently, some have picked up the baton to develop Compound outside the Compound Labs Team. It’s unclear to us what markets it should focus on – perhaps large, blue-chip markets, especially if it can gain some regulatory advantage.

Value accrual 

The DeFi lending original three (Maker, Aave, Compound) are all rejiggering their strategies in response to the shift to a modular lending architecture. Lending against crypto collateral was once a good business, but when your collateral is onchain, your margins will compress as markets become more efficient. 

This doesn’t mean there are no opportunities in an efficient market structure, just that no one can monopolize their position and extract rent.

The new modular market structure enables more permissionless value capture for proprietary bodies such as risk managers and venture capitalists. This enables a more skin-in-the-game approach to risk management, and directly translates to better opportunities for end users, since economic losses will cause much harm to a vault manager's reputation. 

A great example of this is the recent Gauntlet-Morpho drama during the ezETH depeg.

Gauntlet, an established risk manager, ran an ezETH vault which suffered losses during the depeg. However, since the risk was more defined and isolated, users across other metamorpho vaults were mostly insulated from the fallout, while Gauntlet had to provide post-mortem evaluations and take responsibility.

The reason Gauntlet launched the vault in the first place was because it felt its future prospects were more promising on Morpho, where it could charge a direct fee, versus providing risk management advisory services to Aave governance (which tends to focus more on politiking than risk analysis - you try wining and dining a chainsaw).

Just this week, Morpho founder, Paul Frambot, revealed that a smaller risk manager, Re7Capital, which also has a great research newsletter, was earning $500,000 in annualized onchain revenue as a manager of Morpho vaults. While not huge, this demonstrates how you can build financial companies (and not just degen yield farming) on top of DeFi. It does raise some long-term regulatory questions, but that’s par for the course in crypto these days. And moreover, this won't stop risk managers from topping the 'who's set to gain the most' list for the future of modular lending.

Odds & Ends

• U.S. House Approves Crypto FIT21 Bill With Wave of Democratic Support Link

• Block Analitica proposes new interest rate framework for MakerDAO Link

• DOJ charges two brothers with fraud for baiting MEV bots Link

• Maker founder Rune proposes ‘PureDai’ made up of only crypto collateral Link

• EIP-7706 would add a new gas type specific for calldata Link

• ENS aims to launch own L2, likely with zkSync Link

• Token flow regulatory chart in light of recent US progress Link

Thoughts & Prognostications

• The near and mid-term future of improving the Ethereum network's permissionlessness and decentralization [Vitalik]

• Real World Assets [Our Network]

• Decoding the digital dollar [Matías Andrade & Tanay Ved/Coin Metrics]

• New launches (part 1) - private capture, phantom pricing [COBIE/Cobie]

• What Infinex tells us about next-gen front-ends [Peteris Erins/Auditless]

• Big blocks, blobs, and reorgs [Toni Wahrstätter/Ethresearch]

• ETH ETF ramblings [Kelly Ye/Decentral Park Research]

• Verifiable compute: Scaling trust with cryptography [Dmitriy Berenzon/Archetype]

That’s it! Feedback appreciated. Just hit reply. Thanks to Zhev for major help on breaking down the major modular players. So much green in Tennessee in the spring.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO. Some of my compensation comes from MKR, so I’m financially incentivized for its success. All content is for informational purposes and is not intended as investment advice.

Transaction fees are necessary to support even the most basic activities on a blockchain, as they enable users’ transactions to be granted validity and included in a block. The primary purpose of these fees is to disincentivize spamming; it’s also part of the subsidy paid to validators to build/validate blocks. In a sense, these network fees are analogous to rent; users are paying to access a commodity which is finite per unit of time. The commodity in this context is ‘blockspace’, and is exactly what it implies – space on a block.

Here, we assess blockspace on the two biggest smart-contract blockchains, Ethereum and Solana. And as we look deeper, we learn that fee markets – designed both within the protocol and organically from the bottom-up – enable validators to leverage their access to blockspace.

Solana’s fee market was optimized for high performance and designed to avoid the issues that have arisen from Ethereum’s approach. Yet while Solana’s market may ultimately be more performant than Ethereum’s, it still needs to go through a similar MEV revolution to its peer’s (where validators begin to take advantage of their privileged position). Solana doesn’t have to go the same proposal-builder-separation (PBS) route that Ethereum chose, but it will need to settle on a holistic approach to stabilize its fee market long term.

The ABCs of blockspace valuation

Before we dive in, let’s try to understand how blockspace value is roughly determined.

There’s both a technical aspect and a social layer (essentially, the entities whose collective trust in the blockchain affords it value). On the technical side, blockchains can adjust the block size, block time, and block-production and propagation mechanism. See the below chart for a more detailed description and comparison in approach by Ethereum and Solana. 

The social aspect is the coordination of blockchain stakeholders to achieve the chain’s technical and fiscal goals. It could also be viewed as the chain's social status, which is subjective but nevertheless an important measure. Social pressure works, as does setting a particular culture for resolving issues – something Solana and Ethereum have both established. Recent examples of discussions around the social layer are the ongoing arguments over whether to increase Ethereum's gas limit and issuance per epoch, and the recent closure of Jito's mempool on Solana.

Now, let’s move on to a more detailed review and contrast of the fee markets on both Ethereum and Solana. 

Ethereum's fee market: A summary

Ethereum's popularity is mostly due to its execution environment: the Ethereum Virtual Machine (EVM), which enables smart contracts. Another factor is that the permissionless nature of Ethereum has yielded various innovative applications across multiple cycles: the ICO craze of ‘17-'18, DeFi summer 2020, and ‘21-'22s NFT mania. The continuing presence of these applications creates value that is transferred to validators, who provide the blockspace for these activities.

Soon after economic activity proliferated on Ethereum, miners (this was several years before the switch to PoS) began exploring ways to use their position as block proposers to insert their own transactions when arbitrage opportunities arose.

Phil Daian was the first to document this activity – which we now refer to as MEV – in his seminal paper, Flash Boys 2.0 in 2019. At the time, the Ethereum fee market only allowed higher gas prices as a way to incentivize transaction inclusion. These priority gas auctions (PGAs) clogged up the Ethereum network and raised gas prices until Flashbots (co-founded by Daian), launched. This then created a market for miners to get paid for transaction inclusion by searchers, who are on-chain arbitrage traders. Ethereum researchers subsequently realized that MEV extraction could be a more powerful motivator than in-protocol fees. 

Perhaps the biggest change to Ethereum’s fee market was EIP-1559, which created a base fee (dynamically deterministic per epoch, disincentivises spamming, burned), and a priority fee (used to show urgency or specify preferences, and paid to the block proposer for transaction inclusion). An important point to note is that ‘priority fees’ are functionally different from ‘tips’. The former ensures inclusion and is mediated by the underlying chain, while the latter ensures ordering alongside inclusion and is mediated via a fee market.

Ethereum’s approach has continued to evolve; check out our two part deep dive on MEV from last fall. This has occurred through a combination of the social layer that is trying to decentralize a centralizing MEV industry, and the technical layer where MEV is now a key part of the technical roadmap (Vitalik calls this part of the roadmap, ‘The Scourge’).

Solana's fee market: The mechanics

Solana was built with an acutely different approach to blockchain architecture, especially regarding scalability. 

Some of Solana’s notable innovations include:

• The absence of a general mempool: In Solana, transactions are directly forwarded from the originating client to the current leader in charge of producing a block, eliminating the need for a mempool. This theoretically reduces the latency for transaction confirmation, but isn’t always the case in practice due to ‘jitter’, which is the variation in processing times that different validators experience when handling transactions or blocks.

• State isolation: An extension of its lack of a mempool, enabling transactions on its dAPPs to be more independent of each other. This approach is analogous to the ‘adding more lanes to ease traffic’ principle; different types of transactions on Solana have a specific ‘path’ they have to follow from the user to the leader in order to be added to a block. This article provides more detail.

• Parallel execution: Solana's ability to process non-overlapping transactions concurrently in the same block. This is aided by two factors:

  • Block production is (roughly) continuous in Solana, since leaders are expected to add transactions to blocks as they receive them.

  • The slot-leader role is deterministic, as they are scheduled in a queue beforehand, and leaders are also responsible for producing four consecutive blocks.

These two factors, alongside Solana's state isolation, enable transaction ‘multithreading’. This is where the current epoch leader schedules multiple packets of transactions to be confirmed at roughly the same time (on the condition that transactions in the same threads do not alter the same state) in the same way and at the same time.

Solana’s fee market: Cheaper ≠ better 

Network fees on Solana are typically very low (although they’ve risen with recent demand). In contrast to Ethereum, Solana has a static base fee per signature measured in lamports. Its priority fee is then measured in microlamports per compute unit requested.

This means that while fees algorithmically scale with complexity and demand on the EVM, the SVM scales only its priority fees via simple demand. The technical issues arising from this non-dynamism are detailed here, but the gist is that pricing a commodity whose demand fluctuates wildly while its supply is deterministic in a static manner isn't optimal.

Solana’s fee market: The MEV inevitability

The social consensus on Solana is that its low fees are its distinctive advantage over other blockchains. This approach invites spam, so some have called for higher fees or a dynamic base fee for times of high activity (akin to EIP-1559).

Solana’s approach to date has been to implement localized fee markets in response to heightened demand. Since states are isolated, it’s trivial for the network to determine ‘hotspots’ or states that are experiencing a surge in demand. This hotspot-approach enables a blockchain to algorithmically price targeted transaction fees above the average for transactions compared to other less in-demand states. This approach – similar to the block-builder role on Ethereum – is done by a scheduler, which helps to place transactions in continuous blocks based on priority fees. 

As part of the implementation of local fee markets, Solana built an in-protocol scheduler, which natively schedules transactions to be executed based on a first-in-first-out algorithm. Transactions are continuously streamed to the slot leader who then sorts the transactions based on the tips they offer.

The algorithm also requires that the slot leader shares the shreds they are building with some of the nodes they're connected to, based on the latter's stakes. However, as noted earlier, this process is disrupted by jitter. Specifically, scheduler jitter (arising from Solana's random assignment of incoming transactions to execution threads) and network jitter (from delays in P2P relay of incoming transactions and shreds).

These ‘jitters’ contribute to non-deterministic transaction ordering on Solana, which makes blockspace auctions economically viable. So, in other words, whenever there is jitter, validators have an economic incentive to insert or reorder transactions. For users, this means MEV leakage and for validators, MEV profits.

Solana vs. Ethereum

A quick MEV-Ethereum recap: on Ethereum pre-Flashbots, MEV activity crowded out regular blockchain activity, pushing up gas prices for all users through PGAs. On Solana, fees don’t spike because it does not have a shared state and a global minimum price like Ethereum, but it's incredibly difficult for regular users on Solana to land a transaction during heightened activity. Flashbots released MEV-GETH to deal with PGAs, creating a separate lane for MEV value to be auctioned off outside of the in-protocol fee mechanism. In the case of Solana, Jito launched a similar product for validators that gave them access to a psuedo-mempool and a customized scheduler that would order transactions in the most profitable way. Jito’s mempool was attractive to users, offering them guaranteed inclusion for the right to be front-run (aka, have their MEV extracted).

While a popular product, Jito’s mempool came under social pressure and was shuttered last month. This is likely for a similar reason as to why more than 20% of Ethereum transactions run through private mempools: users are tired of being sandwiched. Spamming is now back on the table as the sole mechanism for (probabilistically) guaranteed execution of time-sensitive transactions on Solana. And the absence of a mechanism for efficient blockspace bidding leads to uncertainty during high demand.

Since transactions on Solana are now streamed directly to the slot leader and the prioritization model has been broken, topology (and latency by extension) is the most important component users will consider for time-sensitive transactions. 

A user's topology in the network may be understood as how ‘far’ they are from the leader, and is dependent on the weight of their stake and/or the stake of the nodes they are connected to. Thus, rational agents will seek to be connected to nodes which already control high amounts of stake, leading to centralisation.

As a short-term consequence of spamming, Solana is now so congested that it’s practically unusable for less-savant users because of transaction failures. And addressing the long-term consequences (co-location and centralization of network stake) has become even more important.

A more structured market?

Solana’s initial design philosophy was centered around eliminating user friction and allowing the validator network to meet demand in whatever way it could. What Solana missed was that markets do best when they have some underlying certainty about how they function. Fee markets provide a way to democratize inclusion by requiring users to pay more, moving the problem from a topological perspective, to an incentive-based one. 

While this changes the user experience, embracing fee markets – particularly as they relate to MEV – is the best path forward for Solana and its users. Arguably,providing a cost-intensive avenue to inclusion, while maintaining a chain's integrity, is far better than no avenue at all.

Indeed, onchain activities are almost always time sensitive, especially when an agent seeks to extract value with little-to-no economic cost. Overpriced deterministic execution is better than cheap probabilistic execution.

The specialization of fee markets enables the bargaining and auctions for blockspace to occur at higher levels away from consensus and execution. Validators can thus carry out their duties without having to worry about optimizing for best outcomes to accrued blockspace value.

The impending MEV revolution on Solana

Solana is in the midst of a chain-wide conversation on how its fee markets should restructure (something Ethereum has been pondering for years, yet still not figured out). 

Solana has yet to go through the necessary MEV transition. And while the recent increase in activity on the chain has attracted MEV players like Jito and Ellipsis to begin building MEV infrastructure, major validators have yet to cross the rubicon and start running their own Solana MEV strategies. This is in contrast to all major staking providers on Ethereum that run MEV boost. The Solana validator community is not as adversarial as Ethereum’s, so the handshake agreement to not extract MEV in order to prioritize the end-user experience has been held (so far).

This will not last; the social layer cannot police behavior forever. Blockchains must function in an adversarial environment with self-interested actors. Solana might fare better than Ethereum because it can solve some MEV problems without the heavy decentralization shackles that Ethereum wears. Yet it will have to answer thorny questions like, should all staked SOL get access to MEV rewards like Ethereum has achieved through MEV boost?

Some minimization mechanisms are already being explored with the aim to solve Solana's congestion problem. These include a dynamic fee structure, upcoming changes to the native scheduler specifications, stake-based limitations, and other optimizations at the application level. Things are moving quickly. Jito’s CEO recently acknowledged that a “small cluster of operators/searchers [are] sandwiching [and] running private mempools”.

MEV is a marker for economic growth and as such is unavoidable. In fact, even Bitcoin – whose simplicity is usually hailed as its greatest feature – is beginning to experience a reinvention following the rise of Ordinals and economic activity. Choosing to ignore solutions due to negative externalities (such as in Jito's case) does not eliminate said externality, it simply leads to an uncoordinated market.

The social layer is an effective tool for stopping predatory behavior, but only for a short time. Ethereum is experiencing the shortcomings of the social layer with the rise of timing games, a strategy where block proposers intentionally delay the publication of their block for as long as possible to maximize MEV capture. This weakens the security of the chain but makes economic sense from a validator perspective. Shame works for a little while, but protocol research is the only long-term fix.

It’s too early to say what Solana’s MEV supply chain will look like in a few years' time. But one thing we can be sure of for now is that most value will be captured by large validators.

Odds & Ends

• Bitcoin halving dashboard Link

• Frax Finance expands to Cosmos via Noble Link

• Paradigm open sources Reth AlphaNet, an OP Stack-compatible rollup Link

• Synthetix V3 goes live on Base Link

• Highly anticipated restaking protocol, Eigenlayer launches on ETH mainnet Link

• Alternative Layer-1 Monad raises $225m before launching its testnet Link

• Senators Lummis and Gillibrand introduce new stablecoin bill Link

• MakerDAO raises sUSDe & USDe debt ceiling to $1bn Link

Thoughts & Prognostications

• Blobsplaining Part 2: Lessons from the first EIP-4844 congestion event [Blair Marshall/Blocknative]

• Onchain finance is thriving; what’s next? [Mario Laul/Placeholder]

• Lending protocols [Our Network]

• On ERC-4337, Intents, and MEV [Ben Basche & Alex Watts/Fastlane]

• Solana Uncovered: An introduction to the Solana Network [Caron Schaller/CoinMetrics]

• "We Plan to Win" - Hayden Adams on Uniswap vs. the SEC [Bankless]

• State of DeFi Q1 2024 [Toe Bautista & Kinji Steimetz/Messari]

That’s it! Feedback appreciated. Just hit reply. Great to finally get a deep dive on Solana on Dose of DeFi. I’ll be in Berlin next month at Dappcon & ETH Berlin

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO. All content is for informational purposes and is not intended as investment advice.

DeFi yields have continued to surge higher. Until very recently, the lending landscape consisted of stalwarts like Maker* and Aave printing money with higher rates while being challenged by a new crop of lending protocols. These newbies, the likes of Morpho and Ajna, entered with more market-based risk management – as we discussed in our January examination of the lending landscape. 

That was more or less the state of play until an emergency proposal was submitted to MakerDAO on March 8^th to hike interest rates to insulate against the risk of peg instability. The proposal, which was passed and implemented within three days, tripled borrowing rates overnight and raised the Dai Savings Rate (DSR) to 15% (up from 5.5%).

The drastic measures were in response to the sudden success of Ethena’s sUSDe stablecoin. USDe is currently offering 35%+ yield (past seven-day return annualized), generated by tapping into the basis trade on centralized exchanges. Such yield opportunity only exists because investors are so bullish right now, willing to pay high funding rates to stay levered long.

While there was some initial hesitation, long-time DeFi industry leaders have quickly embraced sUSDe and are in a mad dash to integrate it into their lending protocols. Queue the point where we highlight the glaring truth: things could also go south and lead to a painful unwinding if market sentiment (and the funding rate) change quickly. Lenders are currently working to ensure that such risk is mitigated, but this doesn’t negate the obvious: The Yield Wars have begun. 

DeFi has always been about yield. After TradFi rates shot up in 2022, DeFi finally has competitive yield that justifies the very real risk, and more importantly, will be much more sustainable than the governance tokens handed out in liquidity mining campaigns in 2021. Ethena is tapping into a novel source of yield, tokenizing it and offering it to DeFi degens. This is a playbook that will be repeated in the future as the black hole of DeFi sucks in any available yield.

Crypto innovation before DeFi

Nowadays, all financial innovation in crypto is labeled ‘DeFi’, but the term only started being widely used in 2019. Bitcoin was the OG and what the cool kids call a “paradigm shift”, enabling the digital transfer of value in a censorship-resistant manner. Stablecoins were a further breakthrough in pre-DeFi days. The other big financial innovation in pre-DeFi crypto was the perpetual swap, first developed by Bitmex in 2016.

Perpetual swaps (or perpetual contracts) are similar to a futures contract but do not have a fixed expiry or settlement date. They simply roll over – wait for it – perpetually. They’re also a favorite because they provide easy leverage (Bitmex was notorious for offering 100x leverage). Perpetual swaps are by far the preferred method for going long an asset and using leverage.

These swaps can provide this exposure by maintaining a balance between investors going long and short. This works via a funding rate that incentivizes positions that the system needs to stay balanced. Most perpetual swaps set new funding rates every eight hours. At that point, the system determines the number of longs and shorts and makes a calculated payment to all accounts holding the position it needs to incentivize, and deducts a small amount from those with positions the system has too much of. 

The basis trade

As our readers may be aware, crypto investors tend to be hyper bullish, travel in packs, and love leverage (perhaps, ‘gaggle’ is the appropriate collective noun). When the bulls start running, degens buy crypto perpetual swaps to maximize their exposure. Paying interest on this crowded trade is a small price to pay for being a bull. This market dynamic creates arbitrage opportunities for sophisticated traders.

The trade is simple: you hold BTC or ETH and short BTC or ETH. Or better yet, you find the exchange offering the juiciest yield for shorting and the best yield for longing. This is called a delta-neutral strategy because there’s no exposure to the price of the underlying asset. Since both positions are held simultaneously, the value of your position does not change if the price of the underlying asset goes up or down.

Holding spot and shorting futures (or perpetuals), aka the basis trade, is also a popular strategy in TradFi, where hedge funds will take advantage of the market discrepancy created from pension and long-term investors preferring to hold Treasury Futures rather than spot.

We can tokenize that

Ethena launched as a way to tap into yield from the funding rate on centralized exchanges and offer it onchain. Ethena's CEO Guy Young says he was inspired by Bitmex founder Arthur Hayes’ ‘Dust on Crust’ post about a scalable crypto stablecoin using perpetual swaps. Ethena’s core product is USDe, a token that is pegged to the dollar and collateralized by a delta-neutral position, one long ETH and one short ETH. USDe holders can stake their tokens and receive sUSDe in return, which comes with the yield that Ethena generates with the collateral backing the token. 

The yield passed onto sUSDe holders comes from two sources. First, the long ETH position is in the form of a liquid staking token (LST) which generates 3-4% yield from protocol rewards and MEV. Second, the short ETH position pays handsomely on all major CEXs (upwards of 20-40% annualized yield).

The 20%+ yield on a stable asset without reliance on governance tokens has caught the attention of the market, with many having flashbacks to Luna and perpetual motion machines. Ethena is different. All yield is paid by another investor (or ETH protocol rewards). USDe has seen explosive growth, reaching a $1.4 billion circulating supply just four months after launch.

In total, there is $9.94 billion in open interest on ETH derivatives across CEXs & DEXs, so Etherna is almost 15% of that.

In addition to tapping into a novel source of yield, the other innovation is that USDe can scale relying on crypto collateral with the capital efficiency of USDC/Tether, “ Since the staked ETH collateral can be perfectly hedged with a short position of equivalent notional, the synthetic dollar only requires 1:1 ‘collateralization.’” Its ceiling is the size of ETH open interest, but of course it could also tap into BTC funding, an even larger market.

DeFi’s interconnectedness

Launching a new stablecoin is a fairly common occurrence in DeFi. As one of the most successful launches, Ethena introduces new competitive dynamics in the stablecoin market. Those will play out in the coming months, but the popularity of its product and the investor rush to access it has had repercussions across the DeFi lending space. Insatiable investor appetite for sUSDe led to the most drastic interest rate changes at Maker since Black Thursday in March 2020.

Let’s examine why.

Dai has two different types of collateral backing it, with separate feedback loops that contribute to peg stability, interest rates on crypto-collateralized loans, and real-world assets (including the stablecoin USDC), which does most of the heavy lifting for peg management. Maker introduced the Peg Stability Module (PSM), which allows for a 1:1 conversion of Dai to USDC, shortly after Black Thursday. This has ensured a tight peg for the last four years, but also opened Maker up to criticisms as USDC approached 50% of all collateral-backing Dai. Maker drastically reduced this percentage by taking a portion of the USDC in the PSM and investing it in T-bills through Blocktower and Monetalis, and via a similar arrangement with Coinbase Custody.

These moves gave MakerDAO access to the 4%+ yields in TradFi, increasing Maker’s revenue by more than $100 million and paving the way for an increase in the DSR to 5.5%.

The only downside was that this left less USDC in the PSM. Going from T-bills to USDC is doable in theory, but not on DeFi time. So while real-world assets are great for Maker revenue generation, they make Dai slightly less liquid onchain and increase the chance of peg stability issues.

Maker’s evolving role

Okay, so back to Ethena and the popularity of sUSDe. Its high yield comes from the high interest rates crypto degens were willing to pay to increase their crypto exposure further. But it’s not itself minted from leverage. So, say you have a stack of ETH and you see this 20% yielding stablecoin. You don’t want to sell your ETH, so you use it as a collateral and borrow against it (at an interest rate much lower than sUSDe yield), then buy USDe and then stake it. Ah, the carry trade. What a classic.

Maker has long been the cheapest source of leverage on crypto collateral in DeFi. Unlike Aave, Compound, or other money market-based lending markets, Maker does not require stablecoin deposits to make stablecoin loans. Importantly, many now access this cheap borrowing through Spark, rather than directly through Maker core, although it's the same outcome.

So as a diligent crypto degen, your typical flow is borrow Dai against crypto and either: 1. sell Dai and buy more crypto or 2. sell Dai and buy USDe and stake it to get sUSDe and 25%+. 

The key to Maker is everyone is minting Dai (good) but selling Dai (bad). This has introduced downward pressure on the peg and with the PSM as an easy exhaust valve, the Dai supply has declined over the last two months, bucking the conventional wisdom that Dai circulating supply expands in bull markets because investors take out more crypto-backed loans.

(De)central banking

MakerDAO governance and more specifically Block Analytica was attuned to these dynamics. The growth of sUSDe decreased demand for holding Dai in the DSR, which was only offering 5.5%. While there wasn’t a risk of MakerDAO being insolvent (its collateral is good!), there were significant risks to peg stability with the depletion of USDC in the PSM. And with the weekend approaching (and TradFi markets closed, delaying ability to move more RWAs into USDC), MakerDAO governance acted swiftly to raise interest rates on all crypto-backed loans to 16%+ and set the DSR at 15.5%.

The proposal went up on Friday morning (US time) on March 7^th, was passed 24 hours later, and then executed on Monday, March 10^th after the 48-hour time lock. The sudden moves and overwhelming show of force are reminiscent of the aggressive actions central banks take to calm markets.

MakerDAO’s actions achieved their primary aim: increasing demand to hold Dai, which also led to a replenishment of the PSM with USDC (currently $636 million of Dai’s $4.5 billion supply). And with the DSR at 15%, more Dai was staked to get sDai.

Some protested with the speed of the hikes – wouldn’t signaling aggressive moves be enough? But overall the market has calmed. Rates are expected to come down slightly but – as crazy as it sounds – 15% seems to simply be the ‘risk-free rate’ in DeFi these days.

If you can’t beat ‘em, join ‘em

Crypto lending has come a long way over the past year. The introduction of RWA assets at scale integrated a bottomless source of yield (T-bills) onchain. Getting 5% onchain was great in the bear market, but pales in comparison to the 25%+ yields degens are getting on centralized exchanges. Like RWAs, Ethena created a link between a source of yield and DeFi. There’s a wide design space for how that yield gets translated onchain but in the end, there’s always going to be a token and an APY.

The existing lending incumbents still can benefit from these new sources of yield because they maintain deep liquid markets. Just today, MakerDAO (through Morpho and Spark) allowed up to $100m of Dai to be borrowed against sUSDe after a quick but thorough risk analysis (Aave is considering as well and Ajna already has an sUSDe pool since its permissionless). This could be a new source of borrowing demand and also tighten the link between the funding rate on perpetual swaps and DeFi credit markets. In the future, lending protocols will be the distribution channels for new sources of yield.

Rehypothecation, what could go wrong?

Of course, adding more collateral will introduce more leverage to the system. With sUSDe as collateral on Morpho, all (rational) sUSDe holders will deposit their sUSDe into Morpho, borrow Dai at 16%+ and then buy USDe, stake it and repeat it. This yield optimization strategy is how borrowing rates from CeFi can translate to DeFi.

With high rates and leverage, we should be mindful of how it could all go south. Unlike Luna, a shock to the confidence of the system can’t destroy the underlying value of USDe. There are typical custodial and fraud concerns, and USDe could face its own peg crisis, particularly if there are liquidity issues for an LST, which happened last year with Lido. But the bigger worry is what happens when the source of yield collapses? USDe would still have value, but sUSDe yields would drop to the single digits. Then the unwinding would occur, and while everything appears solvent, major market moves could expose the gaps (think USDC during the SVB crisis) and exploit the weakest link.

 So keep calm and carry on, but secure your flank. The Yield Wars are here.

* Most of my working time is spent contributing to Powerhouse, which is an Ecosystem Actor for MakerDAO. MKR is part of my compensation package, so I have a financial interest in its success.

Odds & Ends

• 1kx raises new $75m fund Link

• Vaults.fyi, which tracks top yields across DeFi, releases documentation Link

• Calldata.pics tracks blobs usage after Ethereum’s Dencun upgrade Link

• Guide to develop dapps on Flashbots’ SUAVE Link

• IDEX releases vision for IDEX token 2.0 Link

• Base surpasses $500m in daily DEX volume for four straight days Link

• Cow-backed MEV Blocker considers adding fee Link

• BlackRock launches onchain fund with Securitize Link

Thoughts & Prognostications

• Everything is a perp [Joe Clark, Andrew Leone, Dan Robinson/Paradigm]

• How ERC-4626 vaults could reach $10b in TVL [Joey Santoro]

• The spectre of MEV on Bitcoin [Walt Smith/CyberFund]

• Why Morpho will surpass Aave & Compound to become the dominant DeFi lending platform [Dan Elitzer/Nascent]

• An analysis of intent-based markets [Tarun Chitra, Kshitij Kulkarni, Mallesh Pai, Theo Diamandis]

• How “Oracle-less Primitives” Are Rearchitecting DeFi [Robbie Peterson/Delphi Digital]

• Ethereum has blobs. Where do we go from here? [Vitalik] + What else could memecoins be? [Vitalik 2]

That’s it! Feedback appreciated. Just hit reply. Written in Nashville, where I love, love, love spring time.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO. All content is for informational purposes and is not intended as investment advice.

Below, Denis returns to take stock on the state of the modular blockchain space. He first looked at the market opportunity in November 2022, and then how rollups were advancing in May 2023. Since then, the pendulum definitely swung towards Solana and the monolithic design, but already so early on in 2024, the modular narrative is oscillating back.

This issue of Dose of DeFi is brought to you by:

There’s a better way to deploy and manage Uniswap v3 positions. Try Oku.trade out today for a better liquidity provision experience. Follow Oku on X.

A monolithic era?

Solana is all the rage these days, and rightfully so. It’s gone from the dark days of the Alameda crisis, to strong price action, and from frequent halts to successfully handling one of the busiest airdrop claims in history – all while maintaining incredibly low fees. From the perspective of onboarding new users, Solana is a good choice: Ethereum L2s still charge up to USD1 per transaction (and we really don’t think starting from BSC or Tron is a good idea).

Another of Solana’s strengths is its single global state that instantly reflects all market signals, without the arbitrage and bridging hops between rollups or shards. It’s as if trading on all global exchanges was seamless 24 hours a day, with events instantly reflected in price changes on all exchanges, no matter the geography or time zone.

These are the benefits of a monolithic chain at its best, but there remain downsides to this design choice. Most notably, that the Solana validator set trends towards centralization due to very high hardware requirements. This happens because Solana monolithically handles all three layers of blockchain: execution, consensus, and data availability. 

On the other end of the design spectrum, modular architecture – and specifically the outsourced data availability layer – is rising in popularity. This approach lowers transaction costs while maintaining low hardware requirements (although MEV threatens this). A modular design also allows for more specialized chains and hardware for specific applications, with dYdX being the best example.

At the forefront of the modular movement is Celestia, a chain optimized for rollup data efficiency. Ethereum, on the other hand, has arrived at a modular approach in a more piecemeal manner, building the plane whilst already flying. We believe rollups are the key to scaling and cheaper transactions, with the battle for data availability layers (and the rest of the modular stack) now on.

Scaling and the data hurdles

The data availability problem was first identified in the early race to scale blockchains. The focus was on minimizing the amount of data to store in order to maximize the number of nodes in a network. The same dynamics underpinned Bitcoin’s block size wars. Data availability refers to a blockchain’s ability to make its data accessible to all network participants. The key breakthrough in solving this problem was the introduction of data availability sampling (DAS), as Bridget Harris explains:

“With DAS, light nodes can confirm that the data is available by participating in rounds of random sampling of block data rather than having to download each entire block. Once multiple rounds of sampling are completed – and a certain confidence threshold is reached that the data is available – the rest of the transaction process is safe to occur. This way, a chain can scale its block size yet maintain easy data availability verification. And considerable cost savings are also achieved: these emerging layers can reduce DA costs by up to 99%.”

Celestia, Avail, NearDA, and EigenDA are the most important DA projects. They don't need to verify transactions, but simply check that each block was added by consensus and that new blocks are available to the network. They rely on third-party sequencers to execute and verify transactions. Celestia was launched in October 2023, Avail and EigenDA have their mainnet in coming months, and Near has most recently announced its DA solution. Let’s review the unique features of each:

1. Celestia chose the fastest road to market with fraud proofs (which are also used by optimistic rollups). The tradeoff is that in the current configuration, Celestia won’t be able to support ZK rollups. Celestia’s team claims that around 70% of all new Arbitrum Orbit chains are using Celestia for data availability.

2. Avail (ex Polygon Avail) as a standalone blockchain provides a fast and secure data and consensus layer that gives developers what they need to launch a rollup (whether a ZK or optimistic).

3. EigenDA is probably the most Ethereum-aligned, because it’s a DA module, not a blockchain. Also, the restaked ETH in EigenLayer will be available to secure rollups using EigenDA. Its weakness is that it’s not using data sampling or data availability proofs.

4. NearDA enables saving on a rollups’ data availability fees, by storing data on Near’s sharded blockchain. NearDA leverages an important part of Near's consensus mechanism, which parallelizes the network into multiple shards.

Launching a sea of rollups

And then we have the rollups themselves. Of the rollups that build on these DA providers, there are a number of tools to make it easier to launch a rollup:

1. By harnessing Celestia's modular data availability, Manta Pacific provides significantly lower costs compared to monolithic L2 solutions, and has already saved $1million in Ethereum gas fees. Manta also uses custom opcodes for verifying ZK technology, which makes it really cheap for them to have privacy and native randomness in the protocol.

2. The Mantle Network is constructed upon a modular architecture that merges an optimistic rollup protocol with the data availability solution from EigenDA. This integration enables the Mantle Network to inherit Ethereum's security while also providing more affordable and readily accessible data availability.

3. Kinto is a KYC’d chain where every user and developer must complete a passport KYC process before transacting on the network. It uses Celestia to drive down costs.

In a truly modular way, modules of each layer are chosen based on specific needs. The variety of combination options can be seen here:

Rollup-as-a-service projects like Eclipse make it even easier to launch a rollup, where the developer chooses which technology to use for each of the three modules. 

Similarly, Conduit allows you to deploy a rollup in 15 minutes, with Optimism, Arbitrum Orbit, and Celestia as the supported stacks. A monthly hosting infrastructure fee is paid to Conduit, and there's a separate data availability fee paid to the provider.

The wealth of possible combinations that modularity creates is certainly a major step forward. Is it akin to the difficulty of building an early website compared to the ease and customization of Squarespace today? 

But then the tradeoffs

Despite the growth in DA projects, many have reservations about outsourcing DA. Vitalik made his clear: “Your data layer must be your security layer.” Dankrad Feist, another member of Ethereum Foundation, concurs: “If it doesn't use Ethereum for data availability, it's not an (Ethereum rollup) and therefore not an Ethereum L2.”

We agree. Rollups with outsourced data availability will be less secure than those using the same chain for data and consensus (and really should be referenced as “validiums”), although secure enough for certain applications. Short-term projects using such rollups will emerge and fade quickly, making it a good experimentation and testing ground. However, for long-term holding of financial assets, L1s such as Ethereum or rollups using them both for data and consensus will remain the networks with the lowest risk profile.

Ethereum is going modular

While skeptical about outsourced data availability, Ethereum is big on modular architecture. The early vision of scaling via sharding was abandoned in favor of modular.

The three main updates needed to implement the vision are rollups (we talked about these before), proposer-builder separation (“instead of a block proposer generating a ‘revenue-maximizing’ block by itself, it delegates the task to a market of outside actors (builders)”), and data sampling. The latter is a way for light nodes to verify that a block was published by only downloading a few randomly selected pieces of data. This is technically more challenging than the other two and will require two to three years to ship. 

Important note: EIP-4844 was the first step in improving Ethereum’s data availability layer before data sampling goes live. As discussed earlier, enhancing Ethereum is similar to building the plane whilst flying; once the Ethereum Foundation recognized the need for rollups (aka when Vitalik dropped the famous rollup-centric future), the team opted to extend blocks with blobs (a dedicated space tailored specifically for rollup data). Blobs are expected to reduce the cost of rollup transactions up to 10 fold. EIP-4844 is scheduled to go live with the Dencun upgrade in March/April. While this is a temporary solution to keep Ethereum competitive for two to three years, the long-term solution will be supporting validity proofs on mainnet itself, which will make rollups orders of magnitude cheaper.

While Solana might be strongly defending its philosophy of monolithic architecture (and they might prove right for many use cases), the industry seems to be converging on modularity. In the case of Ethereum, only modular architecture will enable a future where: 

1. Transactions are cheap for millions of users thanks to rollups (scalability); 

2. The network is protected from censorship and threats like 51% attacks (security); and 

3. An average PC or even a mobile can run a node to verify transactions (decentralization). 

One might ask if Ethereum’s modular architecture solves the blockchain trilemma that was supposed to be unsolvable? Technically it doesn’t, because Ethereum is not a monolithic network anymore, but as a modular network, it does. 

Of these three, we think decentralization is the most important part of the trilemma to solve. Innovation will eventually drive down transaction costs; prioritizing decentralization (especially geographic) is the only way to ensure long-term security for the network. Ethereum is leading in decentralization by having the most distributed validator set, with more than 800,000 validators. At the same time, with the modular approach, it can adapt to new design innovations through customized rollups that launch on top. Celestia and others certainly share this vision. The question remaining is whether Ethereum can move in this modular direction fast enough to keep up with the competition, which is building from scratch, and not fixing the plane whilst flying.

Odds & Ends

• Uniswap Foundation proposes fee switch for token holders in governance Link

• Superstate launches on-chain, regulated fund tracking TBills Link

• DEX aggregator Jupiter airdrops $700m to Solana users Link

• Synthetix 2024 roadmap Link

• Dirt Roads writer launchs M^0, “money middlware for the digital age” Link

• CoW DAO releases MEV-capturing AMM Link

• Lending protocol Morpho sees huge growth Link

• a16z invests $100m in EigenLayer Link

Thoughts & Prognostications

• Leaderless Auction, a decentralized auction with no auctioneer [Dave White & co./Paradigm]

• OFAs are a bad idea [Snoopy_mev]

• How Polygon is changing the game with AggLayer [Peteris Erins/Auditless]

• Introducing the Chain Abstraction Key Elements (CAKE) framework [Ankit Chiplunkar & Stephane Gosselin/Frontier Research]

• Should we be taking Tron seriously? [Michael Nadeau/The DeFi Report]

• Unbundling staking: Towards rainbow staking [Barnabe/ETH Research]

• On Stablecoins [Our Network]

• At the crossroads: Bitcoin ETFs & Coinbase earnings [Tanay Ved & Matias Andrade/CoinMetrics]

That’s it! Feedback appreciated. Just hit reply. I’m in Denver next week, get in touch if you’re around.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO. All content is for informational purposes and is not intended as investment advice.

DeFi lenders came out of the market carnage of 2022 bruised but still functioning (which could not be said for their CeFi brethren). Just as crypto markets were crashing and demand for leverage was plummeting, the Fed added to DeFi lenders’ pain via its most aggressive rate-hiking campaign ever, sapping demand for the modest yields in DeFi. And then in March 2023, the fall of SVB and the depegging of USDC threatened to push all lending protocols underwater.

A year later and that fear has disappeared; DeFi lenders are flourishing. On-chain rates now look more like those in TradFi (thanks to some nudging from Maker), and a whole new crop of innovative products have broadened collateral base and diversified the risk offerings – something that’s certainly set to continue. 

While DeFi trading collectively struggles to offer a product more competitive than CeFi, DeFi lending has already surpassed CeFi and offers more potential to grow into mainstream markets with higher yields. Thanks to higher rates, DeFi lending protocols are seeing an explosion of on-chain revenue that will be reinvested back into expansion into new markets. We don’t expect a bull market sugar high, but rather a shift away from the DeFi of the zero interest rate environment toward a new period of higher revenue and broader appeal. 

Short period, great progress

Rates climbed steadily throughout 2023, driven by two forces: the increased demand for leverage as markets heated up, and real-world yield coming on-chain. MakerDAO led the charge with higher rates when it voted to increase the borrowing rate for its vaults in the spring, even though demand for borrowing was tepid. Unlike Compound and Aave – where interest rates are set by market demand – Maker has full control over what interest rate it issues for Dai loans. 

Maker was not raising rates just to stick it to borrowers. Higher lending rates allowed it to raise the Dai Savings Rate, first to 3.33%, and later to 5% (where it remains for now).

The growth of real World assets (RWA) in MakerDAO brought higher TradFi rates to DeFi and greatly reduced their stablecoin exposure. Dai was sometimes derisively mocked as “wrapped USDC”, but this is no longer the case. At the beginning of 2023, 63% of Dai was backed by USDC or another stablecoin. Fast forward to today, and that number is 12%. It has been replaced by real-world assets (primarily T-bills), which are yielding 5%+. This involves a lot of off-chain complexity swapping USDC in the Peg Stability Module for T-bills and dripping payments back to the protocol in Dai, but it scales better than crypto collateral. 

Throughout 2023, Maker has maintained its lead in the ETH lending market ahead of Aave. In 2021, Aave benefited from an aggressive collateral onboarding strategy. This played out well for them in the transition to staked ETH, and their early mover advantage enabled them to lock in the liquidity network effects. It has more staked ETH than Maker. Staked ETH represents 59% of Aave’s ETH-based collateral, compared with 42% for Maker. Compound missed out on the staked ETH market, but remains a competitive second place (behind Aave) in WBTC.

MKR has led in token performance. Its standout growth, however, occurred over the past few months, on higher revenues from RWAs and speculation about changes in the Endgame (more on that later).

New protocols and innovations

The close competitive dynamics between Maker, Aave, and Compound (and other smaller players) has helped drive a wave of continued innovation and new product releases in lending. This involves both new and old players alike. The most exciting recent developments are as follows:

• Morpho launched in August 2022, and now has a couple of different products. The Optimizer is a P2P lending protocol built on top of Compound and Aave. It matches up lenders and borrowers directly and falls back on Compound or Aave. Just this month, Morpho released Morpho Blue, a protocol with a very thin layer of governance and off-chain dependencies. Instead of relying on core protocol governance to manage lending risks. Morpho Blue is permissionless at the base layer, but lending pools can be curated by DeFi service providers like Block Analitica, Steakhouse, or B. Protocol. It also offers higher loan-to-value (LTV) ratios due to enhanced capital efficiency. 

• Ajna just launched this month. It’s even more governance-minimized than Morpho and also boasts a non-oracle design, as well as being able to accept any ERC20 or NFT as collateral. It uses an internal order book for lending positions, meaning users sets a market price for what their collateral could solve for. Ajna could be thought of as an “interest-bearing limit order on the borrowed asset”. It’s not very normie friendly, but its flexibility means that others might build primitives on top of Ajna to make it easier to access for less sophisticated users.

• GHO & crvUSD are new stablecoins from Aave and Curve respectively. GHO remains small (market cap: $35m) and has struggled with peg management. It’s still early days though, and given Aave’s position in the lending market, it can take its time building up an ecosystem around GHO. Presumably, Aave is interested in replicating the RWA model that Maker has pioneered. That’s not the case for crvUSD. Its strategic advantage is that it makes LPing on Curve more attractive, and there’s some reflexivity to it as a stablecoin hub. Its key innovation its slow liquidations. They just announced additional lending plans this week with code already up on github.

• Spark is a fork of Aave v3 that was launched as a subDAO of MakerDAO. It is plugged into Maker core through the Direct Deposit Module (D3M). Unlike Compound v2 or Aave, the Maker Protocol is not a money market and does not lend out collateral (it uses the CDP model). Spark is a way of offering this product while still connected to Maker’s balance sheet. Spark’s Dai borrow rate is set by MakerDAO governance (currently 5.53%). It offers the best borrowing rates because it’s not dependent on its supply of Dai deposits. It can simply mint new Dai. Spark is just the beginning of some major changes to MakerDAO that revolve around the ambitious Endgame plan. For example, Spark will be one of several “subDAOs” that can borrow on Maker’s balance sheet. These subDAOs will have new tokens to be farmed out, starting this year. Maker is also getting a complete rebrand and a new token. The farming and rebrand should bring much needed attention to DeFi.

Additionally, lending protocols have seen significant growth in deposits and borrowing on their L2 deployments. Token emissions are a big reason for this, but the assets are likely to stay after incentives as interest rates  on L2s are also relatively competitive with Ethereum. L2s and low-cost chains are very important for DeFi lenders’ expansion, as it’s the only way they can move towards a more mass retail investor base.

Swimming with the trends

In DeFi, market dynamics are more fluid in lending, and success seems more dependent on catching the latest wave of investor preference. Compound’s early success was due to COMP farming, which kicked off a wave of copycats. Aave pulled ahead because it cornered the LST market. 

Looking ahead, lending protocols will next be battling to attract the upcoming onslaught of restaked ETH tokens expected to hit the market with the launch of Eigenlayer. We discussed the “ETF-ization of ETH yield” last year and since then, several projects such as Rio, Renzo, and Swell, have already announced plans to launch new forms of tokenized restaked ETH. This will be trickier than staked ETH, where there are different tokens for the same yield (ETH rewards from the Ethereum protocol). With restaked LSTs, there will be scores of different types of yield with different risk considerations. The market will likely be more splintered than vanilla staked ETH, where Lido has 70%+ market share. Most interesting will be to see how conducive lending protocols are to highly-leveraged recursive borrowing, given that the asset prices of the restaked LSTs will be highly correlated, even if their risk is not.

Lending’s crossover appeal

Yield is a powerful motivator for investors. For the crypto-curious, DeFi’s high rates can be easy to write off as coming from “fake governance tokens”. But yield can also be pure, as demonstrated by the appeal of 5% risk free in TradFi last year. DeFi lenders could be the key to onboarding more casual DeFi users if they can offer sustainable yield. A significant amount of capital could flow from CeFi crypto users alone, if yields are attractive.

The most promising aspect of DeFi lenders is that they’re making money. MakerDAO could reach $200m in revenue this year, with expenses under $30m. Aave, meanwhile, may see $35m in yearly revenue, while Compound could surpass $15m. These are going to the DAOs and can be reinvested into growth and protocol development. Allocating that capital to entities that are aligned on a strategy and mission will require effective and efficient governance. Lending protocols have professionalized their risk management through service providers like Gauntlet, Block Analitica, and Chaos Labs, but they’ll need to extend this focus to other governance areas going forward. Lending protocols are complex and require coordination, but the last few years have proven that the model truly can scale and be a sustainable business model.

Odds & Ends

• Bloxroute product to help validators navigate consensus timing games Link

• What it takes to be a block builder Link

• Flash market making MEV protocol Aori launches on Arbitrum Link

• Paypal participates in CRV bribe governance for its stablecoin Link

• Khlani Network aims to be decentralized solver infrastructure layer Link

• Orderflow.art Link

• Optimism RetroPGF Round 3 recipients Link

• Synthetix launches v3 perpetuals on Base Link

Thoughts & Prognostications

• Crypto's Three Body Problem [Toby Shorin, Laura Lotti, Sam Hart/Other Internet]

• Loss versus Rebalancing 101 [Alphaist/Titatnia]

• Evolution of the block building game [Danning/Flashbots]

• A path for sustainable relay incentivization [ballsyalchemist/Fenbushi]

• Auctions: context, types of auctions, what is an auction from the Auction Theory Perspective and from the Mechanism Design Perspective, and EIP-1559 as a specific example [Taiko Labs]

• Spark Protocol: An overview [AXSN]

• Tether’s Ascent: Breaking down the dominant stablecoin's growth [Tanay Ved/CoinMetrics]

That’s it! Feedback appreciated. Just hit reply. Written in Nashville, where I’m still recovering from our big snow last week.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. I spend most of my time contributing to Powerhouse, an ecosystem actor for MakerDAO. All content is for informational purposes and is not intended as investment advice.

Top-five memes

1. Loss versus rebalancing (LVR)

The first on-chain AMM Bancor launched in 2017. This was followed by Uniswap in 2018, and soon after, impermanent loss was coined to describe the risk of LPing in an AMM with sharp (or really any) price movements. Tim Roughgarden of Columbia and a16z crypto reframed the problem and rebranded impermanent loss to loss versus rebalancing (LVR). Eliminating impermanent loss seemed like an impossibility, but reducing LVR feels like a “challenging but well-defined problem”, in the words of Dan Robinson. New research also emerged this year from Max Resnick at the Special Mechanism Group that illustrated the CEX-DEX arbitrage, which some say is responsible for 85-90% of all MEV profits. 

Minimizing LVR has now become a modus operandi for all DEX designers. Uniswap unveiled its v4, with its core concept of hooks giving LPs – or more accurately, new DEX designers – more tools to deal with LVR. This has not yet launched but its code is public. Ambient Finance has a similar concept but launched on-chain without the catchy hook descriptor. It has dynamic fees and hopes to reduce toxic flow. Sorella Labs has yet to release its Angstrom project, but it aims to use a gated liquidity layer built on Uni v4, with the ability to auction off the right to trade with LPs. This is similar to the design that CoW Swap has researched. All focus is now on reducing LVR and incentivizing searchers to work on behalf of LPs.  

Learn more: 

• Ending LP’s losing game: Exploring the Loss-Versus-Rebalancing (LVR) Problem and its Solutions [0xKeyu/Fenbushi].

• Will MEV mitigation destroy on-chain liquidity? [Dose of DeFi]

• Uniswap v4 and Ambient Finance: Fresh hope for passive LPs [Dose of DeFi]

2. (Un)aligned with Ethereum

Ethereum is a complicated web of competing ideas and interests. There’s the Ethereum Foundation (EF) and the social layer, and then there’s application developers and users. Becoming aligned with all of these stakeholders has become a bit of a joke in 2023.

On one hand, you have other chains competing with each other to show how committed they are to Ethereum, like Polygon funding Ethereum core devs to show its zkevm is aligned with Ethereum. On the other hand, some people say they are “unaligned” with Ethereum because they disagree with the technical direction the protocol has taken. This group is critical of PBS and the MEV ruling class it has spurred. Some want faster block times, while others are frustrated with the solo-stakers obsession. 

Explore further: Ethereum, to be, or not to be (aligned)? [Abdelhamid Bakhta/Starknet].

3. The run on SVB and the USDC depeg

At a macro level, 2023 was defined by interest rate hikes at record pace. This wreaked havoc at regional banks and ultimately blew up Silicon Valley Bank and tarnished the sterling reputation of the DeFi’s Blue Knight (USDC). It was fitting that during the frantic weekend before the FDIC announced that SVB depositors would be made whole, USDC was considered the broader market price for an SVB recovery.

USDC is the most popular stablecoin in DeFi lending protocols, so its depegging could have created bad debt from mispriced collateral. On Compound, USDC was hardcoded to $1.00, so there was no risk of bad debt. Since then, Compound released v3, which only uses USDC as its borrowing stablecoin and does not as collateral. Aave, meanwhile, experienced the biggest scare during the USDC depeg. USDC did not drop below $0.85 and thus threaten the large mainnet markets, but Aave v3 with efficiency mode (E-Mode) saw some losses. Maker engulfed an extra $2 billion in USDC, leaving it even more dependent on the stablecoin. The event left such a sour taste that it’s pushing Dai to move away from the US dollar along a path laid out in the Endgame by Maker founder Rune Christensen. It has also onboarded other RWAs to significantly decrease the reliance on USDC. 

Return: Damage control: DeFi lenders and the USDC depeg [Dose of DeFi]

4. Intents

When you first hear “intents” in a serious crypto conversation, you probably wonder why there needs to be a word for it. But then you see how the word has been used as a self-descriptor for a swath of new “intent-based protocols”. Could these exist without first the creation of the intents meme? Was CoW Swap an intent-based protocol when it launched before the meme was popularized? Put on your tin foil hats everyone.

In reality, the shift in user behavior from one where an execution path is defined to one where an end goal is signaled is subtle but important, and necessitates a word to describe the new design space. Looking ahead, application developers will build for an intents world, but there is still a lot of debate on how these intents will be solved. Front-ends and applications are where MEV is leaked, so they will need to come up with solutions to control their MEV supply chain.

Background: Intent-based architectures and their risks [Quintus Kilbourn & Georgios Konstantopoulos/Flashbots & Paradigm]

5. Restaking and consensus overload

Restaking is one of those ideas that is so intuitive, you wonder how it was not thought of before. Put simply, a restaking protocol enables ETH validators to generate additional yield by running other computational tasks, such as an oracle. Sreeram Kannan theorized the idea that became Eigenlayer at the University of Washington in 2021, but it was this year that the idea truly broke through to the mainstream Ethereum conversation. Eigenlayer raised $50m in March. Its testnet launched in April and it went live on the mainnet in June, supporting stETH and rETH. Already, $400 million has been staked on EigenLayer, reaching the cap for stage 1 of the mainnet launch, even though there is no explicit yield promised. 

Restaking changes Ethereum in two key ways. First, it offers a credible alternative to projects flirting with the app chain model of Cosmos. Cosmos’ appeal is the ability to easily bootstrap a small validator set that does more than just consensus, but actually runs logic on behalf of the application. dYdX has already shown the technical advantages in this approach with their v4. The rising popularity of coprocessors will also be a key use case for restaking.

Second, it introduces more leverage to Ethereum. This is what Vitalik is talking about when he wrote about not “overloading Ethereum’s consensus”. The same assets are being rehypothecated to secure more and more economic activity. Vitalik’s concern is how a downstream problem for a heavily restaked protocol might threaten Ethereum’s credible neutrality.

Restaking opens up the design space for trustless compute but it also introduces more layers of risk that must be monitored.

Dig deeper:

• Restaking and shared security - the future of blockchain infrastructure [Felix Lutsch/Chorus One]

• After Shapella, a new dawn for ETH yield products [Dose of DeFi]

• Restaking Alignment with Vitalik, Sreeram, Tim Beiko, Justin Drake, Dankrad & Jessy [Bankless]

Top-five charts

1. Lido’s continued dominance

Lido has become perhaps the most important protocol in DeFi after it cornered the market on liquid staking tokens (LST’s) in 2022. There has been considerable anxiety on the power that Lido has. It has maintained a big lead on the LST market with over 75% market share, and a considerable lead in the overall ETH stake market hovering around ~30%. Many in the ETH community see keeping Lido below the 33% threshold as key to maintaining Ethereum’s decentralization, so it is notable that it did not significantly increase its market share, although the overall amount of ETH staked increased by 80% this year with ~23% of all ETH staked. 

What are the downsides of Lido’s dominance? In their own words: "The most important concern is probably the validator/operator whitelist. If Lido continues to gain market share, there’s a risk that LDO holders would be able to effectively determine the majority of the Ethereum validator set. Governance could then shepherd Lido’s operators into working together to exploit multi-block MEV, execute profitable re-orgs, and/or censor certain transactions." In February, Lido announced their v2, intended to decentralize the node operator list.

However, amidst these risks, Lido provides a buffer against KYC-ed exchanges dominating the Ethereum validator set (Coinbase is catching up with 15% share). We anticipate intense discussions about the platform's impact in the coming years.

2. LSTs take over ETH as primary lending collateral

DeFi lending protocols did not unveil new protocol designs like their DEX brethren. Instead, they were focused on diversifying diversifying its collateral base. The most significant change of the year was the shift from ETH as the primary source of collateral to LSTs, specifically Lido’s stETH, which now accounts for a staggering 56% of collateral in major DeFi lending platforms. This explains Lido’s rise and reinforces its dominance, as its liquidity in DeFi simply makes it more useful than other LST’s.

Elsewhere in lending, Real World Assets (RWAs) increased dramatically, most notably at MakerDAO. Up until USDC’s depeg, almost 50% of Dai was backed by USDC. Maker onboarded almost $2.5bn in RWA collateral in the form of US Treasury bills, facilitated by BlockTower and Monetalis Clydesdale, which has decreased its reliance on centralized stablecoins to just under 9%. Maker also launched Spark, another lending protocol that forked Aave v3, which experienced rapid growth over the second half of the year.

3. Rollups proliferate

In 2023, rollups emerged as a pivotal force in the industry. From the earlier discussions to the proliferation of diverse rollup solutions, the sector experienced a significant surge. Total value locked (TVL) has shown consistent growth, fostering healthy competition among ecosystems. Users now have the autonomy to opt for chains offering lower transaction costs, which are already below Ethereum's rates (though still not reaching ideal levels). However, as the user base expands, transaction batching will reduce costs.

What interests us most is the diversity in the rollup offering (see chart). Optimistic rollups currently dominate. Arbitrum (after launching an airdrop in 2023) is leading in TVL, with Optimism following. However, the latter is growing horizontally – Coinbase’s Base chain is using Optimism Stack and is a part of its Superchain. Even though Vitalik predicts the eventual triumph of the ZK technology, ZK-based zkSync, Scroll and Starknet are yet to gain significant traction (and yet to launch tokens).

4. Tether increases its lead

Stablecoins have endured the bear market quite well, with total value slowly decreasing from $140 billion at the peak of the bull market to $130 billion at present, significantly less than the 40% decline in the total crypto market cap. Despite this, USDT has maintained its leadership position, witnessing growth in both absolute value (from $65 billion to $90 billion) and in market share (from 48% to 70%). Meanwhile, USDC backed by Circle appeared to be making all the right moves, from closer collaboration with regulators, to expanding onto more chains and developing a top-tier bridge. Yet its market share declined from 31% to 19% in 2023. The USDC depeg is the most likely culprit for the decline. Tether's USDT has firmly established itself within exchanges and is gaining momentum in South-East Asia and Latin America. Does USDT’s “up only” performance show us that the stablecoin market ultimately favors a winner-takes-all scenario?

5. Number go up…

Comparing prices at the end of the year is always an exercise in the arbitrary but it’s still fun none the less, especially comparing how the coins are doing at the end of 2023 compared to 2022. Who knows if this is the beginning of a sustained bull market, but it’s clearly put some animal spirits back into the industry.

The coins have all gone up, but not at the same pace. Solana is on an absolute tear, especially over the last two months, reminiscent of how ETH outperformed BTC in 2021. More recently, Solana is starting to see some uptick in DeFi activity. BTC, meanwhile, has reminded us to respect the orange coin. It has such a unique position in the public psyche. It seems some normies have reclaimed the narrative from the maximalists and there is now stuff actually being built on Bitcoin.

ETH, meanwhile, has lagged behind. It has long been the center of DeFi, so we are clearly biased towards it. Two problems hang over the ETH narrative: inability to scale and MEV extraction. These are only problems for Ethereum, because it’s the only blockchain with enough activity to create these problems in the first place. An army of researchers and developers are trying to solve these problems now. What’s clear is whether it’s a multichain world or a modular one, there will be not be one chain to win them all.

That’s it! Feedback appreciated. Just hit reply. Written in Nashville, where I’m fully embracing the holiday season. May I be one of the first to say, “Happy New Year!”

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. All content is for informational purposes and is not intended as investment advice.

This issue of Dose of DeFi is brought to you by:

Oku makes DEX trading seamless across the top EVM chains. Try it now with 0% fees and advanced analytics. Website & Twitter

DeFi has to date been dominated by investors with either niche technical know-how, or enough hubris to fly blind. Yet when trading goes beyond a browser plugin to institutional levels, a myriad of issues arise. The reality is, DeFi has evolved for a market of individual traders managing their own capital, but operational challenges arise when “OPM” comes into play. Since the beginning of financial markets, and across all asset classes and market cycles, traders have tended towards using more and more OPM. For those unfamiliar with the term, OPM lovingly stands for, “Other People’s Money”. This article is for traders and potential investors looking to evaluate the current landscape of institutional capital in DeFi. We will not be focused on market speculation, but instead survey the challenges traders and investors face today and how it impacts you.

Here’s an executive summary:

• There are roughly four institutional ways to custody in DeFi:

  1. Hardware wallets

  2. Smart contracts and trading bots

  3. CeFi’s DeFi integrations

  4. Simulation-based approaches

• Simulation based approaches scored the highest across our metrics, while CeFi’s DeFi integrations appear to pose real and poorly understood threats to their clients. This was our most surprising discovery.

• DeFi is still maturing. There is a small but growing industry of professional traders and fund managers in DeFi. If DeFi is to live up to its potential, this crowd will become very important. We believe this is worth paying attention to.

Separation of church and state

Over the last hundred years, traditional markets have evolved to enforce the separation of trading and administrative privileges at every layer of the organization. Furthermore, large legal and technical systems detail precisely what the rules are when it comes to any given financial product or service. These modern structures protect investors from excessive financial risk, internal collusion, theft, and other malfeasance. 

Over the last few years, CeFi has started to trend in this direction (albeit through bankruptcy and arrests) and as regulation comes into play, one may expect the market to largely replicate this familiar model. Crypto is a different asset class but the underlying market structure is comparably similar: centralized ownership of assets, operated either on-premise or on cloud-hosted services, where the speed of trading and the security of the assets is ensured by a few institutional operators.

The non-custodial nature of DeFi, however, makes managing capital at scale a more challenging problem. If a “large” DeFi fund (which today would constitute assets on-chain of only $40-100M) wants to actively trade its book, it runs into challenges around custody, trade execution, and data integrity. We explore these nuances in detail below. By and large, the central problem is how a fund manages its transaction security, which in CeFi is encompassed by custody of the assets, but in DeFi takes on a broader context. In assessing their options, DeFi fund managers - and perhaps more importantly - their LPs are affected by a tradeoff between transaction security, automated execution (e.g. a stop-loss button), and the ability to dynamically adjust risk. 

DeFi’s core ethos is to build a base financial layer with open and equal access for all investors. In order to grow and reach mass adoption, DeFi will need to serve professional fund managers, who serve you and I, and bring market efficiency and much-needed liquidity.

It all starts with custody transaction security

When Bitcoin emerged from the ashes of the Global Financial Crisis, one of its more popular memes was to “be your own bank”. Through public key cryptography, blockchains enable a string of characters (the private key) to unilaterally control an account (the transaction outputs of a public address). The idea is summed up in the phrase “your keys, your crypto”. With Bitcoin, the only thing to do is hodl, which doesn’t spur much of a conducive financial system. DeFi picked up where Bitcoin left off and facilitated trading, lending, and other financial services with self-custody – trusting only smart contracts to execute predetermined trade logic.

Allowing everyone to be their own bank means that everyone also needs to store their own keys. Storing a private key on a phone or computer is fine if there are only a few hundred dollars in the account, but the calculus changes if that number is $10m or more.

Until recently, the only solution for large investors came in the form of centralized custodians that look and feel like a more traditional SaaS or financial firm. Custodians such as Coinbase, Anchorage, and Paxos will safeguard a private key and come bonded and insured. The issue here is that these custodians are primarily designed to hodl, and so they don’t realistically allow their clients to participate in DeFi. 

For investors who want on-chain exposure, there are four main custody options, as shown in the table above. From this, it’s clear that there’s no perfect solution as they all involve tradeoffs between private key security, automated execution, and the ability to easily modify trading strategies.

Hardware wallets: On the solo scale

Individual traders are typically comfortable using a hardware wallet, like a Ledger, and storing the private key someplace safe. The benefit is that private keys are offline, so even if a device was compromised, no trade could be executed. 

Hardware wallets are highly flexible in their ability to interact with any DeFi protocol and on almost any chain. They typically ensure good private key security, because the private keys are not easily compromised. The downside is they’re not scalable and most importantly - humans don’t read EVM bytecode, which has led to the long list of hacks and theft headlining search results. Still, this may be a functional setup for a small-ish DeFi fund doing mostly simple swaps or yield farming. Using a Gnosis Safe with multiple hardware wallet signers adds redundancy, but also makes it difficult to act quickly, and doesn’t solve the core problem of screening for potentially malicious transactions (for which there are solutions described below). Importantly, multi-signature wallets enable only a half-solution to the problem of separating administrative and trading privileges. 

Trading bots & smart contracts: Scalable, flexible, and secure… kind of

While some DeFi funds may be content with swapping and yielding, others are running more complex strategies across multiple protocols and chains. Human signatories cannot be relied upon here. In the time it takes to initiate and sign a transaction, the opportunity has likely moved on or the damage is done.

Instead of humans, bots running on servers execute predefined trading strategies dependent on various market conditions. This is what most MEV traders do. For instance, a bot could be running a just-in-time (JIT) liquidity strategy on Uniswap v3, where it monitors the public mempool and immediately supplies liquidity when it observes a big swap, to earn the LP swap fees. To do this, the bot server needs to store the private keys, meaning whoever has access to the server has access to the keys and all the funds it controls. 

To solve this access problem, firms write smart contracts that restrict the total functionality of the contract custodying the assets. Consequently, even if a private key was compromised, a malicious actor could not steal or redirect the funds to its own address.

This approach has historically been the only realistic option for automated trading. While it sufficiently protects the private key (or more accurately, removes the singular dependence) and enables real automation, it has one major downside, namely, firms need to write, test, and deploy a new smart contract for every adjustment in the trade, resulting in two prohibitive problems:

1. Hedge funds, whose survival is predicated on reacting quickly to market conditions, are slowed to the speed of an engineering team who isn’t allowed to make mistakes.

2. It is prohibitively expensive to secure the long tail of smart contracts, and as a result, it typically isn’t. There are regular instances of MEV bot smart contracts getting exploited. 

In essence, it is kicking the private-key-can down the proverbial smart-contract-road.

Automated trading systems are essential for most professional fund managers. Yet problems arise when automated trade execution meets custody. One potential workaround being explored is the use of CeFi custodians to manage private keys for DeFi funds.

CeFi’s DeFi integrations: Most popular, most expensive, and least secure

The most popular option for large DeFi fund managers comes in the form of a crop of CeFi custodians that offer DeFi integrations. These service providers’ core products are their custody solutions (typically multi-party computation or MPC wallets), OTC trading, and CeFi integrations. They offer a predefined policy engine that manages risk and allows fund managers to give certain permissions to different users on their team. 

These CeFi custodians can be divided into three different groups. 

1. The first offers the most vanilla on-chain services, like staking and on-chain governance. They’re firmly rooted in a “safety first” approach, but at the cost of minimal functionality. Anchorage Digital is the best example. 

2. The second offers DeFi integrations through Metamask Institutional or some other browser wallet. Using these custodians – Bitgo, GK8 and Qredo, among others – is potentially useful for a fund that is doing basic DeFi activity, like yield farming, swapping, or lending but doesn’t expect to need more than a small handful of functions. 

3. The last group of custodians – best exemplified by Fireblocks, Cactus, and Copper – brand themselves as essentially “DeFi native” firms. They advertise a number of flexible services, including a configurable policy engine and automated execution for DeFi strategies. This hypothetically allow programmatic access to on-chain contracts and code which can set triggers for customized liquidity management, trade execution, or exit strategies.

The third group is the most important, as it advertises the functionality that is necessary to  trade professionally on-chain. In order to prevent malicious activity, these services apply a policy engine that whitelists certain smart contract addresses that traders are allowed to interact with. The problem is that while they advertise features such as the ability “to deploy systematic DeFi strategies while maintaining the highest level of fund security on an institutional-grade platform” and an API “that enables programmatic access to smart contracts, while extending security to every DeFi interaction" their policy engines do not actually check the behavior of on-chain transactions - neither for manual nor automated trading. 

These firms only check high level ‘to’ and ‘from’ fields of a DeFi transaction, ignoring its behavior (encoded in what is called the “calldata”). This approach is the security equivalent of asking for one’s DOB on certain adult websites... Consequently, firms and their investors are often under the impression they are being protected from theft or effectively separating trading and administrative privileges when they in fact are not.

This vulnerability indicates that these firms are adding DeFi functionality to an existing product, rather than building a DeFi-native system that understands the nuances of how blockchain transactions work. However, there is an emerging industry of DeFi native providers that have one important thing in common.

Simulation-based approaches: DeFi native, but not a silver bullet

Over the last two years, DeFi native startups tackling “the transaction security problem” have evolved into more dependable service providers. There are, so far, three groups of solutions, all with one thing in common - they all take a “transaction simulation based approach”.

Simulating the transaction allows either a person or a policy engine to look at the result of a transaction and judge whether it is secure. For example, if as a result of the transaction, funds end up in an account you’ve never seen, no matter how it happened, you likely want to reject that transaction.

Where these firms differ, is their approach to custody and private key storage. There are roughly three categories:

1. Custodians - Fordefi is a direct competitor to the likes of Fireblocks, Cactus, and Copper for their DeFi business. Unlike the CeFi custodians, their policy engine is based on transaction simulation. The upside is they credibly protect their clients in DeFi, in contrast to the aforementioned custodians. The simple downside is that most firms already rely on a custodian and changing can be a big headache.

2. Security analytics solutions - Examples include Pocket Universe for individuals and Hypernative, Redefine, Hexagate, and others for institutions. These solutions provide their clients with visual queues before a transaction takes place, allowing clients to avoid high risk transactions. These firms, in contrast to the custodians, do not manage any private key material, making them more of a “security advisor” than a custodian.

3. Co-signers - DeFi Armor (disclosure, built by Eulith) may offer the best of both worlds, but are also the newest of these three categories with DeFi Armor being perhaps the only product in this niche sub-industry. As is the case with the above two categories, they offer a simulation-based policy engine. The difference is in private key storage - their clients can choose their own custody solution and then separately “plug in” this co-signer, which stores an additional private key and rejects transactions automatically if they are unsafe. 

While our research indicates simulation-based approaches are the best we have, they’re not a silver bullet either. There are two main downsides to be aware of:

1. A transaction simulation can take up to several seconds, which is too slow for certain high-frequency strategies. In these instances firms are back to rolling their own smart contract security.

2. A simulation-based policy engine is not inherently bulletproof. As with any security system, there are ways to get it wrong. The most common way is ignoring the potential consequences of pre-trade state-change (a topic for another article!).

The bottom line is while simulation-based approaches appear to be the best, institutional firms should test these solutions before depending on them for large allocations.

DeFi is still maturing

We see the future of financial systems in DeFi because of the implications of self-custody, inherent transparency, and permissionless access. We’re concerned with maintaining a fair playing field, which motivated our research on MEV. DeFi’s non-custodial design actually gave individual investors a head start; even with the juicy yields of DeFi summer, the custodial options were not robust enough to justify the risk for fund managers. However, this is starting to change, and will be a huge net positive for the industry. 

To accelerate this change, and to help DeFi to scale, the advancement of infrastructure specialized for investors to use is the next critical step. There’s currently a lot of focus on developing better wallets for retail users with social recovery, but what’s equally needed is a robust way for institutional investors to access DeFi without compromising risk management. Importantly, these innovations are being built on top of blockchains, and don’t require a compromise on DeFi’s commitment to a permissionless financial system.

Special thanks to Moh Rezaei and Kristian Gaylord for feedback and review. Special thanks to the many dozens of firms who gave us their valuable time and insight in developing our research.

Odds & Ends

• Kyber responds to hackers’ absurd demands Link

• Eden releases public datatsets on block building and OFAs Link

• Yearn launches v3 on Polygon Link

• 15% of Ethereum tx flow through private mempools, 50% of non-toxic flow Link

• Flashbots co-founder launches Alfred, a Telegram trading bot Link

• US House of Representatives hopes to pass stablecoin bill in early 2024 Link

• Major proposals in November to MakerDAO’s protocol parameters Link

• Over $600m flows into multi-sig for Blast, a new L2 with native yield Link

Thoughts & prognostications

• Similarities and differences between OFAs and PFOFs [bloXroute Labs]

• A brief intro to coprocessors [Emperor]

• Deep dive into Kyber exploit [Doug Colkitt/Ambient Finance]

• The impact of Uniswap X on the DEX ecosystem [Orange Finance]

• Overview of different approaches to rollup sequencers [Arixon]

• My techno-optimism [Vitalik]

• Switching gears: A step towards sustainable credit [Saurabh/Decentralised]

• Eyes and Orders: On Crypto Front-ends [Trace/Figment Capital]

That’s it! Feedback appreciated. Just hit reply. Delayed post because of Devconnect in Istanbul. Written in Nashville. I’ll be in NYC next Wednesday & Thursday at Columbia’s CryptoEconomics summit. Holler if you’re around.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. All content is for informational purposes and is not intended as investment advice.

As we discussed last month, the MEV ecosystem is growing rapidly. Instead of trying to solve MEV, focus has shifted to incorporating it into application and protocol design. MEV mitigation and democratization have become the most important things in DeFi (and arguably, all of Ethereum). Yet while MEV mitigation efforts are necessary to make Ethereum easy to use, the solutions emerging over the past year (while competitive), have come at the expense of decentralization. 

This has led to soul searching over the true goal of the DeFi ecosystem. Is it creating products that are competitive with TradFi and centralized crypto exchanges? Or is it achieving some undefined decentralization end point?

This debate is best exemplified by the recent rise in order flow auctions (OFAs), the MEV mitigation strategy that returns the value extracted back to the end user (for a fee).

There are many different solutions in the works (and we discuss the main ones later on), but they all follow the same pattern:

• Order routing and execution are delegated to specialized networks outside of Ethereum.

• This intermediate computation layer relies on trusted actors to mediate between users and validators.

• These new networks empower sophisticated (and centralizing) market makers, but this comes at the expense of on-chain liquidity. 

Is non-custodial trading with professional market makers inevitable for DeFi? Decentralizing the MEV supply chain and democratizing MEV rewards are hurdles to clear, but this will not be the catalyst of new innovative DeFi products that conquer CeFi and TradFi. 

Bypassing the dark forest

Let’s first do a stock take on the more recent approach to MEV mitigation. Over the past year, there has been a huge shift in how DeFi users trade on Ethereum. Increasingly, they are sending their trades not to the public mempool but to private networks that fill and execute orders and return a portion of the MEV extracted back to users. These OFAs, where searchers/solvers/fillers bid on the right to execute orders using both on and off-chain liquidity, come in all shapes and sizes

These networks serve as trusted (for now) intermediaries. On one side are retail users and the other is a set of market makers, bidders, searchers or solvers, which we’ll refer to as facilitators. Shifting order routing off-chain to facilitators yields a significant gas cost savings, while users get better prices because facilitators can also use off-chain liquidity. Facilitators could theoretically trade against their users (frontrun or sandwich) but this would jeopardize their future flow. So, it’s in the best interest of the OFA networks to ensure facilitators are not acting maliciously. 

OFAs have different ways of policing facilitators. Some only allow certain entities to bid for order flow (permissioned), while others allow anyone to participate but can retroactively punish bad actors (either by slashing a stake or just banishing them from the network). Others allow facilitators to bid for orders permissionlessly by using a batch (CowSwap) or Dutch auction (Uniswap X, DFlow, and 1inch Fusion). This requires additional time for the auction to complete and the OFA network must still trust that the block builder is not extracting MEV when it passes along the transaction bundle to validators.

There’s now a feverish rush to redirect all retail flow from the public mempool to private OFA networks. Uninformed retail traders are the source of MEV, so anyone that controls that flow can attract the best facilitators and market makers to their network. OFA networks are targeting wallets and RPC providers, as well as building their own front-ends. Several prominent organizations have launched their own private mempools, like Blocknative’s Transaction Boost, Flashbots Protect, and MEV Blocker (from Gnosis). Meanwhile, a number of independent transaction-based private mempools, such as Merkle and Kolibro, have recently emerged. Owning retail flow is the most important part of the MEV supply chain.

OFA networks: Intents vs generalized transactions

In evaluating the rise of these OFA networks, we also need to remember a key differentiator:  between those whose input is the user's transaction, versus those that are a user’s intent. Intent is a made-up term – crypto is very good at that – to describe a user’s desire for a certain outcome, rather than a transaction, which is specific lines of code to execute. Flashbots’ Quintus Kilbourn and Paradigm’s Georgios Konstantopoulos define it succinctly:

If a transaction explicitly refers to “how” an action should be performed, an intent refers to “what” the desired outcome of that action should be. If a transaction says “do A then B, pay exactly C to get X back”, an intent says “I want X and I’m willing to pay up to C”.

For most users interested in executing a swap at the best price with the lowest fees, using an intent-based architecture is ideal. Most users don’t care how their order is executed, only that whoever executes it abides by conditions defined by the user. 

This harkens back to the classic debate on whether blockchains are best thought of as a world computer or a global settlement layer. Clearly, there are benefits to smart contracts and trustless execution, but at what cost? It also fits into the trend of moving execution to Layer 2s in general. The difference being that when using Arbitrum or Optimism, you are using the same execution engine as Ethereum (the EVM) and posting a proof of the execution on Ethereum. OFA networks and intent-based blockchains (like SUAVE, Anoma, or DFlow) will also need to offer similar trust assurances, if they want to maintain the same trust assumptions as execution on Ethereum. 

The diminishing value of on-chain liquidity

Efforts to mitigate the negative externalities of MEV via OFA networks have had one obvious casualty: on-chain liquidity. All efforts to route trades to private OFA networks mean that facilitators have the first look at filling orders. Facilitators can tap into liquidity offered by on-chain automated market makers like Uniswap, Balancer, and Curve, but they will only get flow when facilitators decide it’s not worth using their own liquidity. On-chain liquidity providers are already getting slammed by CEX-DEX arbitrage and problems from loss versus rebalancing (LVR), but in the new market architecture to minimize MEV, they will also forfeit their exclusive access to retail DEX traders.

Liquidity providers strike back

So, are on-chain liquidity and MEV mitigation really mutually exclusive? Not everyone thinks so. Current efforts to minimize MEV extraction while still empowering on-chain liquidity providers (LPs) are underway. The simplest method is to return MEV from CEX-DEX arbitrage back to LPs. The process goes like this: A price change on Binance leads to a rush to trade against Uniswap LPs that have not yet incorporated the new price. There’s a huge amount of value to be extracted from whomever gets to trade against the LPs first. The McAMM (MEV capturing AMM) design would auction off the right to the first trade per block and return the proceeds to LPs.

Empowering LPs is also the foundation for Uniswap v4 and Ambient Finance, which we called “Fresh hope for passive LPs” in a deep dive in June. Both of these AMMs enable pools to be programmed with “hooks”. The design space for these has not yet been fully explored (Ambient launched in June and Uni v4 has yet to fully launch), but they could impose restrictions on who can trade with a liquidity pool. Theoretically, they could only allow trades from certain OFA networks or private mempools. This means retail flow could be aggregated and only given to specific Uniswap/Ambient LP pools. More importantly, hooks could be designed to protect LPs from toxic flow but it also might limit arbitrageurs rebalancing the pool. Just a few weeks ago, Arrakis released Diamond Hook, a Uni v4 hook meant to alleviate LVR concerns for LPs, although some question whether it can achieve its mission. 

What’s it all for? Why DeFi matters

The efforts to resolve the MEV crisis are rejiggering the entire DeFi landscape, so it’s worth taking a step back to ask, what are we trying to achieve? We’ve been writing about DeFi for almost five years because of the earnestness of those working in it to build a better financial system (despite a lack of consensus on what this actually is!)

To us, DeFi has three value propositions that make it a financial system that conceptually doesn’t enshrine market winners as long-term rent extractors:

• DeFi is transparent. FTX couldn’t happen on-chain because DeFi is non-custodial. BlockFi couldn’t happen in DeFi because the loan book is open and liquidations are automatic.

• DeFi is global first. National regulators can restrict access at the application layer, but DeFi’s default setting is open access around the world.

• DeFi unlocks new financial products and innovations through programmable money and creates an open platform for others to permissionlessly build on top of, much like the internet.

Does the existing MEV supply chain maintain these principles? It’s too early to tell, but there are some red flags. OFAs do not (yet) offer the same transparency as on-chain execution. Facilitators are becoming more sophisticated and it’s possible that we will end up in a world where fillers/solvers/searchers are all large Wall Street market-making firms. This is not inherently a bad thing, but industry concentration makes it an easy target for regulators and threatens DeFi’s global nature. 

As for unlocking new financial products and innovation, OFAs are focused on best execution and the current structure seems an awful lot like the one Robinhood and Citadel already pioneered. 

Even if this became the end state for DeFi, three huge innovations would still have been unlocked:

• Non-custodial trading: The original motivation for creating decentralized exchanges was to allow users to maintain control of their crypto and protect against the risk of hacks, inside jobs, and self-dealing at centralized crypto exchanges. An intent-based DEX settled on Ethereum fulfills that vision.

• Bootstrapping liquidity for long-tail assets: Even if sophisticated market makers are the primary liquidity providers for blue chip assets, on-chain AMMs are still the best way to bootstrap liquidity – and a valuable innovation for the long-tail of assets. Those around in pre-Uniswap times will remember how hard it was to build a liquid market on primitive DEXs like Ether Delta. Permissionless on-chain liquidity pools are an infinitely better option than the shady business practices of CEX listings. 

• On-chain lending pools: The lending side of DeFi has not been as obsessed with MEV as DEXs. While there are issues to overcome (namely around oracles), on-chain credit markets are just getting started, and have continued to produce new, interesting products even in the bear market.

DeFi’s future, sans MEV

Lots of promising MEV mitigation strategies are emerging, but what ultimately wins out is less important than moving past MEV as the focal point of DeFi. The key consideration for any blockchain’s MEV ecosystem is whether MEV extraction or redistribution threatens the three DeFi value propositions we outlined above. Centralizing elements in the MEV supply chain will imperil all three.

So who’s doing it well? dYdX and the Cosmos ecosystem are using the appchain model and on-chain governance to ensure that their validators are not extracting MEV – or at least are transparent about it. This means the chains can stay focused on building great DeFi products. Skip Protocol is a sovereign MEV infrastructure provider underpinning most of the advances in the Cosmos MEV ecosystem.

The rise of OFAs is a necessary step in protecting DeFi users against the dangers of the dark forest. We are most encouraged by SUAVE, Flashbots’ new blockchain (set to launch a Devnet in Q4), because it offers better transparency in execution, privacy protection, and a credible path to decentralizing block production. Decentralized sequencers for Layer 2s, like Espresso or Astria, are not too dissimilar. 

Ultimately, we drink the kool-aid on intent-based systems, and think the delayed auction format used by CoW, Uniswap X, DFlow, and 1inch Fusion is the best market structure to prevent centralization of facilitators while remaining permissionless. Transparency in intents execution is still sorely needed.

Decentralized OFAs will not be enough to beat CeFi. New primitives are needed. The most promising development in MEV mitigation and DeFi innovation more broadly is the new DEX design of Uniswap v4 and Ambient Finance. These are the first DEXs to be designed with MEV mitigation strategies. The concept of “hooks” allows LPs to discriminate between toxic and non-toxic flow, which could finally generate positive returns for LPs.

MEV remains an obstacle to overcome

Many DeFi users were turned off by the rise of MEV due to the constant fear of getting taken advantage of. This rightly drove research into mitigating problems for the average user. The all-encompassing nature of MEV has made it feel like the only game in town. It’s incredibly important to fix, but new types of MEV extraction and redistribution won’t be what drives innovation in DeFi, nor what makes it successful long term. The key to any MEV solution must be one that allows for DeFi to return to its roots: a transparent, global financial system where entrepreneurs can build new financial products that are unlocked through smart contracts and programmable money.

To put our cards on the table, we strongly believe that DeFi will only work long term if there is transparency in how orders are executed, and if retail users can earn some positive return by LPing. Of course, sophisticated actors should be able to earn a higher return, but DeFi will have failed if it can’t produce more than an updated version of an on-chain index fund. Others contend that “passive LPing will [never] be profitable” and the goal is to merely return value leaked to LPs to make healthy markets on-chain. In this world, the fees typically paid to Blackrock would be paid to rebalancers in the MEV ecosystem. So DeFi is having your ETF managed by a decentralized market of traders settled on a blockchain.

These are different visions for the future of DeFi, and now that MEV mitigation and redistribution strategies are in place, we will see which one wins out. Our take:

Odds & Ends

• PoolTogether launches v5 Link

• Ethereum foundation gets $9k extracted on on-chain ETH trade Link

• Ethereum mempool transactions now on Dune Link

• Elixir, DeFi protocol aimed to increase orderbook liquidity, raises $7.5m Link

• Over $500k in cumulative fee revenue for Uniswap labs from front-end fee Link

• USDC supply dropped 44% since ATH to $23bn Link

Thoughts & Prognostications

• Ending LP's losing game: Exploring the Loss-Versus-Rebalancing (LVR) problem and its solutions [0xKeyu/Fenbushi Capital]

• The real fork [Gwart/Doors like this]

• Bite by bite: NounsDAO’s natural demise [Luca Prosperi/Dirt Roads]

• State of the on-chain derivatives market [Saurabh/Decentralised.co]

• Ethereum and Solana [Joel Monegro/placeholder]

• The essential guide to general and app-specific order flow auctions [Xin Wan & Robert Miller/Bell Curve podcast]

• Protocol and staking pool changes that could improve decentralization and reduce consensus overhead [Vitalik]

• Inside Frax Finance: Designing the end game stablecoin [Sam Kazemian/0xResearch]

That’s it! Feedback appreciated. Just hit reply. Special thanks to Daniel MacLennan for discussion and feedback on today’s post. Written in Nashville. Happy fall y’all.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. All content is for informational purposes and is not intended as investment advice.

This is part 1; check out part 2, Will MEV mitigation destroy onchain liquidity?

Ahoy! Back after a restful summer and now drinking from the MEV firehose. There’s so much happening; it reminds me of DeFi in the fall of 2019. Look out for another MEV piece next month.

- Chris

A lot has changed in the three years since Flashbots released MEV-Geth in its attempt to “frontrun the MEV crisis”. Ethereum has been through a market boom and bust, and successfully shifted to Proof of Stake – a major technological feat. All the while, the battle for the value that leaks from economic transfers on blockchains has grown more intense. Attempts to eliminate MEV have failed; the reality is that block proposers will always have an incentive to take advantage of their privileged position. 

The allure of MEV profits now threatens to engulf Ethereum’s sovereignty and censorship-resistance. A central overarching concern remains: that the hunt for MEV will centralize stake in Ethereum. If only the most sophisticated actors running validators are able to reap the rewards of MEV, then ETH holders will gravitate towards these validators. This is because they can offer higher yields with MEV rewards on top of the protocol-enforced yield. 

This centralized future has, at least for now, been avoided through MEV-Boost, which gives large staking pools and solo validators equal access to the fruits of MEV extraction. Yet MEV-Boost was always intended as a stop-gap solution. And while it has generated a robust MEV ecosystem of professionalized players, the industry is becoming increasingly centralized and relies on trusted parties and a single software client.  

We argued last December that “the optimal solution to the MEV crisis will be a standalone, decentralized network focused entirely on sequencing transactions.” However, this is still a (slow) work in progress. Changes to the core Ethereum protocol to enshrine Proposal Builder Separation (PBS) are needed, yet that will not be enough.

We remain confident that the solution to the MEV crisis lies outside of Ethereum, but are now convinced that it will not be a single monolithic solution for addressing all MEV leakage. Instead, different decentralized but vertically-integrated MEV supply chains will emerge that specialize in extracting MEV for specific applications. We also contend there is too much focus on MEV from CEX-DEX arbitrage, which currently accounts for ~60-70% of MEV volume and profits. If DeFi emerges as the foundational global financial market, this won’t be the biggest MEV problem faced. 

MEV: The current landscape

PBS is the overarching design philosophy that ensures that Ethereum remains decentralized and neutral. It’s a widely accepted precondition to any MEV resolution. Ethereum currently achieves PBS through MEV-Boost, the Flashbots-provided software run by validators that allows a randomly selected block proposer to auction off the right to build the most profitable block to the highest bidder. So far, PBS has democratized MEV rewards by allowing solo stakers to partake without employing sophisticated MEV strategies of their own. Yet the rest of the MEV supply chain remains fraught with centralization and censorship concerns. 

As a refresher, the diagram illustrates the major components of the MEV supply chain, a term first coined by Stephane Gosselin, founder of Frontier Tech and formerly of Flashbots.

The Ethereum protocol’s architecture is intended to be much simpler. The design naively assumes that for every block, a randomly selected validator will locally build a block sifting through the public mempool for transactions submitted by users with the highest gas fee. 

The three additional players in the MEV supply chain – searcher, builder, and relayer – coordinate with validators through MEV-Boost, which is currently run by 93% of Ethereum validators. Of the three, the builder is the most prone to centralization, while the relayer is the least rewarded. Searchers used to be the prototypical shadowy super-coders, but now they’re teams of developers: some truly anonymous while others are major trading shops. The economic relationships between searchers, builders, and relayers is shrouded in mystery. Payment for flow is common, but since these relationships are off-chain, they’re not observable.

Increasingly, there are structural advantages for vertically-integrated builders. Just this week, one of the major block builders, Blocknative, announced that it would stop serving as a trusted relayer. It cited the costs of running a relayer (reportedly $500k a year) without any associated revenue. This could have made economic sense if Blocknative had its own team of searchers, but as a US-based company, it’s understandably straying away from any activity that could draw the ire of regulators. 

Bloxroute, also a major builder, has not turned their relayers off, but it’s also hedging its regulatory bets. It runs two relayers, one dubbed “regulated”, which censors blocks with OFAC-sanctioned addresses, and the other “max profit”, which….doesn’t. With the exit of Blocknative, there are now only four major relayers: Bloxroute, Flashbots, Ultrasound (headed by Ethereum researcher Justin Drake), and Agnostic (from the Gnosis team).

CEX-DEX arbitrage: The worst of MEV?

Now that we're up to speed on the current supply chain and its challenges, let's flip back to the mechanics of MEV extraction - and why it's causing such a ruckus. To many, MEV is just frontrunning trades, or worse, sandwiching them (trading the transaction before and after a transaction in the mempool to lock in profits). This is unequivocally bad for regular users. In its most productive form, MEV is on-chain arbitrage between different DEXs. A trade goes through a Uni v2 pool pushing a token price up or down, and mighty MEV bots compete to rebalance other liquidity pools and incorporate the new token price. These are the easiest MEV examples to understand, but they are not the most common.

In fact, the majority of MEV extraction is CEX-DEX arbitrage. This MEV hurts DeFi liquidity providers, not traders. The deepest market for ETH is not on Ethereum: it’s on Binance. And when the price of ETH changes on Binance, there’s a mad rush to trade against Uniswap LPs, who have not yet incorporated the new price. The winning MEV must be the first transaction in the new block after the price change, which is referred to as being “top of block”.

After a significant price change on Binance, an MEV bot must bribe – ahem, pay – the lucky validator who is randomly selected to propose the next block. Of course, in the world of PBS and MEV-Boost, the MEV bot sends its transactions first to a block builder to fill with other transactions, who then pays the lucky validator (as long as they propose the suggested block). All of this happens within 12 seconds (the time in between blocks on Ethereum). Max Resnick of Special Mechanism Group (SMG) explained the CEX-DEX arbitrage in detail at Flashbots MEV Salon in Paris as well as in the full technical paper.

The troubling conclusion: when there is volatility on Binance (a 1% or more change in price), the most sophisticated builders always win the bid for the next block because they’re willing to pay more to be in the lucrative top-of-block position. This creates a circuitous cycle. Sophisticated builders pay more when there is an MEV opportunity in the top of block, meaning they win more of these juicy blocks. This makes them more likely to get private order flow from searchers, and with more private order flow, they can bid more to get their block included. Or, as Max from SMG puts it, “Winners have incentives to get better at winning”. 

This is a latency war with little societal benefit. Yet unfortunately, it’s damn near impossible to prevent. The early bird catches the worm; the most sophisticated players will always get there first. There is considerable research going into solving this problem, and the current consensus solution is to allow for auctions of partial blocks.

We think these research efforts are noble but misguided. DeFi’s end game is not to play second fiddle to TradFi. It’s hard to imagine now, but the whole point is that price discovery should not occur on a centralized exchange. 

Of course, there will always be market-moving information from the off-chain world, where being the first to trade on-chain comes with some advantage. The most obvious example is the Fed announcing interest rate changes. It will never be on-chain first. Looking ahead, MEV in the future will not look like MEV that is extracted now. MEV “solutions” should be generalisable and not tailored to the current problems of the day. Afterall, if Ethereum succeeds, won’t Ethereum be where price discovery for ETH is?

Whack-a-mole

It’s incredibly hard to see the whole picture when it comes to MEV – it’s a dark forest! Still, when it comes to data, EigenPhi has some of the best MEV visualizations. The chart below shows which actors in the supply chain are earning the most profits (relayers would be $0).

Most interestingly, EigenPhi identified that a significant portion of the MEV rewards for validators occur outside of MEV-Boost, meaning that some validators are already developing their own MEV supply chain separate from MEV-Boost. It would be fine if these were solo stakers, but the more likely scenario is that large staking pools are beginning to run their own internal MEV strategies. If true, this could set-off an upward spiral, where higher profits produce better yield, which attracts more and more stake, allowing the staking pool to pay more for private order flow.

Enshrining PBS: It won't be enough

Much like how MEV can be extracted from transactions sent to the public mempool, MEV opportunities could also be sniped by entities within the MEV supply chain. A searcher who discovers an MEV opportunity sends it to a builder in return for a share of the profit. The builder then packs the juicy bundle into an entire block and pays the winning validator to propose the block. In the current design, nothing stops the validator from simply copying the searcher’s transaction submission and replacing it with its own, cutting out the searcher and builder. The reputation of the relayer is what prevents this alpha sniping.

There is no crypto-economic guarantee because PBS is achieved outside of the protocol through MEV-Boost. There has been considerable research on how to enshrine PBS (ePBS) into the Ethereum protocol so there would be no need to trust a third-party relayer to facilitate the builder payment to the block proposer. This is technically challenging and would require changes to the Ethereum protocol (but likely not consensus). PEPC (Protocol-Enforced Proposer Commitments), developed by Barnabe of the Ethereum Foundation, is the most developed example, but we’re still at least 18 months away from possible implementation.

This design does remove the need for a relayer – by creating a neutral way for builders and block proposers to transact – but it will not create an economically-viable competitor to a trusted relay. As Mike Neuder explains in an ETH Research forum post, relayers would still be superior for high-value blocks, like CEX-DEX arbitrage, while also allowing cancellation support.

The one thing that ePBS would improve is censorship resistance. Many of the designs feature the use of an inclusion list of transactions that must be part of the next block. So even if the largest builders were all heavily regulated entities, they couldn’t collude to exclude OFAC sanctioned addresses, for instance.

The best solution will depend on the source of MEV

There’s no shortage of solution ideas when it comes to the MEV crisis. Yet too often, they’re presented as a silver bullet that fixes all problems at once (ahem, fair ordering). We believe that any resolution to MEV extraction must start at the app design phase. Most good app developers are already trying to minimize MEV, but they should realize that there will always be some value leakage. MEV cannot be eliminated. Instead, they should be proactive about engaging directly with MEV supply chain players. Payment for order flow is not an inherently bad thing – only when it’s hidden.

Next month, we will look at the parties taking the lead in MEV supply chain engagement, including Uniswap, Cowswap, SUAVE and Bloxroute & Ambient’s smart routing.

Odds & ends

• Coinbase goes to Washington Link

• Friend.Tech Analysis (MEV, volume, revenue, impact on BASE) Link

• Overview of the RWA asset landscape Link

• Connext launches airdrop and new constitution for DAO governance Link

Thoughts & prognostications

• How ERC-4626 vaults compare across different chains [Vikram Arun/Superform]

• Regime change in the stablecoin market [Nic Carter/Castle Island]

• The engineer’s guide to blockchain finality [Benjamin Samuels/Trail of Bits]

• Operating manual for stablecoin issuers [Steakhouse Financial]

• There are now two types of PayPal dollars, and one is better than the other [JP Koning/Moneyness]

That’s it! Feedback appreciated. Just hit reply. Written in Nashville, which is in the first week of a new mayor! So interested in MEV these days.

Dose of DeFi is written by Chris Powers, with help from Denis Suslov and Financial Content Lab. All content is for informational purposes and is not intended as investment advice.