Open Jupiter, express your desire to exchange SOL for USDC, verify the price, and initiate the trade with a simple click. From the user’s perspective, this process seems straightforward and uncomplicated.
It is common to assume that Jupiter is the platform where the trade actually takes place. However, Jupiter functions more like a liquidity search engine. Its software scans various platforms willing to engage in the trade, compares their offerings, and directs the order towards the most favorable route it can find. Currently, Jupiter’s system can select from decentralized exchanges, proprietary automated market makers (propAMMs), and a request-for-quote network where professional market makers compete to provide a quote.
While the user interacts with a single interface and sees a single price, beneath the surface, various types of markets compete to fulfill the same order.
The entity selling USDC to you might not be another user or a public pool supported by numerous individuals. It could be a professional trading operation utilizing its own funds and pricing software, with no visible consumer exchange or pool for depositing tokens. Most of this intricate process remains hidden from the user, who only sees the final price.
Research from DWF Ventures suggests that propAMMs contribute approximately 15% to 27% of daily on-chain DEX volume. For trades involving SOL-to-stablecoins routed through Jupiter, DWF estimates their share to be over 90%.
These statistics come with a disclaimer: DWF Ventures is part of DWF Labs, a crypto market maker and investor. Nonetheless, the significance lies in what these numbers reveal rather than the exact percentages provided. A form of trading that was previously inconspicuous in the DeFi landscape has now become substantial enough to influence some of the busiest markets.
This scenario places crypto in a peculiar position. The initial decentralized exchanges were established on the premise that markets could become more accessible by substituting professional traders with public software. Anyone could view the pool, contribute capital, and trade based on predefined rules within the code. However, professional traders are making a comeback, and they appear to excel in some of the most liquid markets.
This shift carries more weight as Nasdaq, the London Stock Exchange, Robinhood, and Kraken are exploring ways to migrate traditional stocks onto blockchain platforms.
For years, the financial industry debated whether Wall Street would transition to crypto infrastructure. Now, a more challenging question arises: what type of market structure will be in place when this transition occurs?
The dealers are reemerging in the trading domain. The original automated market maker was a peculiar innovation in DeFi.
Imagine a currency exchange booth with two currency stacks, but with no dealer present. One stack holds SOL, and the other holds USDC. Software connects the two stacks through a formula. Users can deposit SOL and receive USDC, and as the balance shifts between the stacks, the price adjusts accordingly. Additionally, users can contribute funds to the booth and earn a portion of the trading fees.
This setup addressed a significant issue in early DeFi. A new market didn’t need to rely on a professional firm to facilitate trades continuously. All it required were tokens, a smart contract, and individuals willing to invest capital, allowing the market to operate autonomously.
However, self-operation doesn’t equate to efficient trading. If the pool price for SOL is $100, and news triggers a price surge to $101 on centralized exchanges, the pool may lag behind. Arbitrageurs exploit this time gap by purchasing discounted SOL and continuing to do so until the pool price aligns with the market value. The pool, unaware of the price shift, undersells SOL, learning its true value only after offering it at a lower price.
This dilemma, termed loss-versus-rebalancing, highlights the issue of a passive pool trading at outdated prices against informed traders. While this flaw is manageable in stable markets where trading fees compensate for liquidity provision, it becomes costly during rapid price fluctuations.
In contrast, a propAMM operates differently. Instead of relying on external token deposits, a professional trading firm utilizes its inventory. Its proprietary software monitors external markets and adjusts its buying/selling parameters accordingly. While the trade settles on-chain, the pricing mechanism occurs internally within the firm.
This model resembles an electronic currency dealer embedded within a blockchain network rather than the vending-machine concept that characterized early DeFi.
This explains the inclusion of the term “proprietary” in propAMM. Solana’s propAMM explanation indicates that current propAMMs are generally closed-source, their liquidity isn’t accessible for public deposits, and inclusion in an aggregator’s network is a selective process. Users can verify on-chain transactions without inspecting the system that determines the pricing or the rationale behind selecting a specific market maker.
While this deviates from the traditional DeFi ideology, from a user’s perspective seeking a quick and cost-effective SOL swap, the practicality outweighs the ideological debate.
Jump Crypto’s analysis of around 20 million propAMM fills from March revealed that the median SOL-USDC fill deviated by 0.72 basis points from the benchmark centralized-exchange midpoint. Additionally, 91.9% of fills were cheaper than Jump’s estimate for the lowest institutional centralized-exchange fee tier.
Jump Crypto, being a trading firm engaged in this market structure, has its vested interests in conducting the study. Nonetheless, the results elucidate why these venues are attracting orders. Routers direct volume towards propAMMs because, for specific trades, they offer exceptional execution.
While this marks a clear division in the mechanics and economics of decentralized trading, it also signifies a departure from early DeFi culture. The initial DeFi version emphasized openness as an integral aspect of the product itself. The pool was public, the code was transparent, liquidity originated from users, and the capital-to-transaction pathway was visible enough for technically proficient individuals to comprehend most of it.
The latest iteration poses a consumer-centric question: if another system provides a better price, how much of the mechanism must the user witness? Once the answer veers towards “not much,” the exchange starts fading behind the user interface.
Solana aptly accommodated this model due to professional market makers valuing historical prices. A firm quoting SOL at $100 wouldn’t want that figure lingering when the market price is $101. Every additional moment offers traders an opportunity to exploit the price differential before the firm can rectify it.
In comparison to Ethereum’s 12-second time slots, Solana achieved its 300-millisecond target on Aug. 28, as explored by CryptoSlate through the lens of stale-price arbitrage. While the networks vary in transaction processing mechanisms, the allure of Solana is evident. Professional traders aim to update prices swiftly and affordably, and Solana’s emphasis on shorter slots minimizes stale-price risks for liquidity providers and market makers.
Another approach to reach the same outcome is by requesting quotes from professional firms elsewhere and executing only the most favorable trade on-chain.
This concept is central to request-for-quote systems. On 0x’s RFQ network, an application solicits private off-chain quotes from professional market makers, compares them with public AMM liquidity, and selects the route offering the optimal outcome. For extensively traded pairs like USDC-WETH and WBTC-WETH, 0x reports that its RFQ liquidity surpasses AMM pricing about 52% of the time.
Jupiter amalgamates various versions of this concept on Solana. Public DEX liquidity, propAMMs, and professional quote systems operate behind a unified interface, eliminating the need for users to select a specific market type before initiating a trade.
This shift marks a profound transition. Old DeFi essentially instructed users to trade against a public pool. New DeFi is pivoting towards a user-centric approach: convey your requirements, and let the system determine the most suitable entity to fulfill them.
Traditional finance already follows this model. When a user submits a buy order for a stock, the customer typically doesn’t choose the precise market center or dealer executing the trade; it’s the broker who routes the order. The SEC’s explanation of US equity execution delves into brokers selecting exchanges, electronic networks, market makers, or internalization based on the best available execution.
PropAMMs aren’t merely replicas of Wall Street dark pools transplanted onto Solana; perceiving them in that light overlooks the unique attributes of the crypto version. Blockchain settlement is observable, enabling asset movement across public infrastructure, and post-trade verification by anyone. What remains confidential is the mechanism surrounding the transaction: pricing models, inventory decisions, certain route aspects, and at times, the identity of the firm competing for the trade until execution occurs.
This divergence in transparency delineates a new market landscape. Financial markets can expose ownership and settlement for public verification without divulging every aspect of price formation. This distinction assumes greater significance as assets traversing these systems extend beyond pure crypto assets.
Nasdaq recently announced a $100 million investment in Payward, Kraken’s parent company, to collaborate on Nasdaq Equity Tokens. The framework’s launch is anticipated in the second quarter of 2027, intending to integrate with Payward’s xStocks network.
Similarly, the London Stock Exchange is aligning its operations in this direction, with plans to list xStocks for trading on its LSE 24 venue in 2027, subject to regulatory approval. These developments underscore the tangible progress in tokenized stock trading at scale, while addressing the intricate ownership nuances that accompany assets resembling shares.
Robinhood’s blockchain initiative, Robinhood Chain, has transformed tokenized stocks into assets viable in permissionless crypto markets. However, the professional liquidity layer in this setup diverges significantly from the open AMM experiment. 0x elucidates that its Stock Token implementation links applications to professional market makers via RFQ, with USDG as the primary base pair and Tokka Labs as a key maker participating through this system.
This scenario accentuates the critical query that transcends the tokenized finance domain: when traditional assets migrate to crypto platforms, which market structure will they adopt?
CryptoSlate has observed a similar dichotomy in tokenized equities, where markets marketed as middleman-free still rely substantially on conventional financial infrastructure, including brokers and custodians safeguarding underlying shares. One broker now oversees the backing for a sizable portion of tokenized equities. The propAMM transition echoes this broader narrative from the trading perspective: blockchain can reshape the public-facing facets of finance without necessitating the elimination of every intermediary.
The trade is public, but the trading desk remains concealed. The conventional narrative would suggest that DeFi aspired to distance itself from Wall Street, only to inadvertently replicate its functions. However, the actual trajectory may involve disentangling two elements that were once intertwined: settlement transparency and execution transparency.
Blockchains can publicly affirm the occurrence of a trade, transfer assets, and enable asset movement through programmable infrastructure without mandating the public disclosure of the market maker’s pricing algorithm. While settlement transparency is upheld, execution transparency may be confined to private firms vying for trade execution.
There is an economic rationale behind this stratagem. Professional market makers adept at shielding themselves from stale prices and arbitrage exploits needn’t embed these losses into every user price quote. Enhanced information and swift repricing can render a private dealer more cost-effective to trade against than a fully public pool.
The trade-off translates into a shift of influence. As users relinquish the task of selecting exchanges and opt for routers instead, the router assumes a more pivotal role in the market. It determines which firms can compete, which liquidity sources receive orders, the accompanying information, how quotes are evaluated, and how transactions are processed on the blockchain.
Solana’s propAMM introspection highlights several facets of this challenge. Closed-source code, permissioned inclusion by aggregators, and users’ inability to independently verify the winning market maker or the rationale behind their selection until execution transpires are central to this evolving landscape.
As market-making, routing, transaction delivery, and block generation converge, the interplay among these layers becomes increasingly pertinent. Blockworks Research uncovered instances where price updates from HumidiFi reached Harmonic validators sooner than other setups, reflecting the nuanced dynamics in decentralized finance market structures.
This episode serves as a precursor to an impending DeFi market structure discourse: in a multi-layered environment where diverse strata influence execution, what defines best execution on a blockchain?
Traditional finance has spent years formulating regulations around variations of this conundrum. Concepts like best execution, order routing, market access, quote priority, and conflicts between brokers and dealers emerged due to the complexity beyond the initial “Buy” command. DeFi is rediscovering some of this intricacy while reconstructing the financial system using distinct building blocks.
The difference lies in the ability to segregate certain components that were traditionally housed within a single financial institution and distribute them across public networks. This restructuring renders the market more transparent in some segments and less so in others, presenting a nuanced outcome that extends beyond a binary assessment of transparency.
This evolution doesn’t signal the demise of the original AMM. Its inherent vulnerability is also the reason it is likely to persist.
Professional market makers excel in assets with well-established markets boasting deep liquidity. Traditional AMMs retain their edge in niche markets where decentralized market creation holds more value than marginal execution improvements.
Tokenized finance might bifurcate along similar lines. Assets like Apple shares, Treasuries, SOL, ETH, and other heavily traded assets naturally attract sophisticated dealers competing behind routing mechanisms. Assets lacking robust external markets may still rely on public pools willing to facilitate trades with any deposited asset.
Crypto embarked on a grand cultural mission: software would eliminate financial intermediaries. However, the trading markets are now painting a more intricate narrative. While certain intermediaries stemmed from outdated infrastructure, others served a crucial function in pricing risk, managing inventory, and executing trades.
Blockchain technology doesn’t mandate the elimination of this role to revolutionize the surrounding market. The rails can be public, settlement can be transparent, assets can traverse applications instead of being confined to a single institution, and professional dealers can vie for the same trade without owning the underlying infrastructure. The user executing a swap may remain oblivious to the ultimate winner.
The future exchange might embody a blockchain open for public scrutiny, enveloping a concealed trading desk invisible to most.
