💰 quantitative finance

The Viability of Blockchain Markets under Discrete Clearing and Paid Priority

This paper argues that the combination of discrete block-time clearing and paid-priority transaction ordering in blockchain markets creates endogenous participation cutoffs that intensify with competition, ultimately undermining price discovery, exacerbating adverse selection, and potentially causing markets to shut down.

Original authors: Agostino Capponi, Álvaro Cartea, Fayçal Drissi

Published 2026-05-19

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Agostino Capponi, Álvaro Cartea, Fayçal Drissi

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: A Traffic Jam with a Pay-to-Win Lane

Imagine a bustling city where everyone wants to get to a specific destination (buying or selling an asset). In a normal city (traditional stock markets), traffic flows continuously. Cars move in and out, and prices adjust smoothly as new information arrives.

Now, imagine a different city: The Blockchain City.

Discrete Clearing: Instead of moving continuously, traffic stops completely every 12 seconds (the "block time"). Everyone waits at a red light. When the light turns green, a "Builder" (the traffic controller) decides who gets to move first.
Paid Priority: To get through the intersection quickly, drivers can pay a toll. The more you pay, the closer to the front of the line you get. This is the "Priority Fee" (or gas fee).

The authors of this paper ask a simple but profound question: Is this Blockchain City a good place to do business? Their answer is a cautious "No." They argue that the rules of this city actually make it harder to find the true price of things and can cause the market to shut down entirely.

The Three Main Problems

The paper identifies three specific ways the "Blockchain City" design breaks the market.

1. The "Rich Only" Club (Endogenous Selection)

In a normal market, if you have a slightly good idea about a stock's value, you can trade. In the Blockchain City, it's different.

The Analogy: Imagine a race where the prize is getting a better price on a ticket. To win, you have to bid on a "fast lane" ticket.
The Problem: If you only have a small idea that a ticket is worth a little more, you aren't willing to pay a huge toll to get to the front. But if you don't pay the toll, you get stuck at the back of the line. By the time your turn comes, the price has already been pushed up by the people who paid to go first. You lose money.
The Result: Only the people with the biggest, most aggressive ideas (the highest valuations) are willing to pay the toll. The "moderate" traders stay home.
Why it matters: This creates a distorted view of the market. The price doesn't reflect what everyone thinks; it only reflects what the most extreme, aggressive buyers think. It's like trying to guess the average temperature of a room, but only the people wearing heavy winter coats are allowed to speak.

2. The "One-Shot" Gamble (Liquidity Issues)

In traditional markets, if a big buyer comes in, the market maker (the person providing the liquidity) can adjust their price slowly over many small trades.

The Analogy: Imagine a lemonade stand. In a normal town, customers come one by one. The owner sees a few people buying, realizes demand is high, and slowly raises the price.
The Problem: In the Blockchain City, the lemonade stand owner has to set the price before the light turns green. When the light turns green, a massive wave of aggressive buyers (who paid to get to the front) all rush in at once.
The Result: The lemonade stand owner gets hit with a huge, sudden loss because they didn't know how big the wave was. To protect themselves, they decide to provide less lemonade (liquidity) or close the stand entirely.
Why it matters: Because the "moderate" traders are gone (see point #1), the remaining traders are very aggressive. This makes the risk for the lemonade stand owner so high that they pull out, leaving the market with no one to buy or sell from.

3. The "Long Wait" Trap (Block Time)

The time it takes to build a block (the "block time") is necessary to keep the blockchain secure and decentralized. However, the paper argues that making this time longer actually hurts the market.

The Analogy: Imagine the traffic light stays red for a very long time.
The Problem: The longer you wait, the more time there is for the "real value" of the asset to change wildly. By the time the light turns green, the uncertainty is huge.
The Result:
1. Extreme Bidding: Because the value could be anything, the people who do show up are the ones betting on the most extreme outcomes. They bid even higher tolls to ensure they get in.
2. Fewer Traders: The "moderate" traders are even more scared to participate because the risk of being last in line is too high.
3. Market Collapse: Eventually, the tolls get so high and the risk so great that no one wants to run the lemonade stand, and the market shuts down.

What the Data Says

The authors didn't just build a theory; they looked at real data from Ethereum (specifically Uniswap). They found two things that match their model:

Huge Trades: The average trade size on the blockchain is massive (around $70,000) compared to traditional exchanges (around $1,200). This proves their theory: only the "big guns" with huge valuations are willing to play.
Position Matters: The further back in the line (queue position) you are, the smaller your trade volume. This confirms that traders with smaller valuations are being squeezed out or are choosing not to participate.

The Bottom Line

The paper concludes that while blockchain technology is great for security and decentralization, its current design for financial markets is flawed.

The Trade-off: The features that make blockchain secure (waiting for a block to be built, paying for priority) are the exact same features that destroy market efficiency.
The Verdict: If a blockchain is the only place to trade an asset, it will likely fail to discover the true price. It will be biased toward extreme values, lack liquidity, and could collapse if the "block time" is too long or competition gets too fierce.

To make blockchain markets work, the paper suggests we need to rethink how transactions are organized and prioritized within a block, rather than just relying on the current "pay-to-win" queue system.

Technical Summary: The Viability of Blockchain Markets under Discrete Clearing and Paid Priority

Problem Statement
This paper investigates the fundamental viability of blockchain infrastructure as a sole venue for price formation in financial markets. While blockchain technology aims to provide a decentralized alternative to intermediaries, its operational mechanics differ significantly from traditional continuous trading markets. Specifically, blockchains clear transactions at discrete intervals (block time) and order transactions sequentially based on discriminatory priority fees (paid priority) paid by traders to secure queue position. The authors ask whether these structural features—discrete clearing and paid-priority ordering—undermine the market's ability to discover prices and provide liquidity, potentially leading to market shutdown.

Methodology
The authors develop a theoretical framework modeling a three-stage game involving a representative liquidity supplier (providing depth to a Decentralized Exchange or DEX), informed traders, and noise traders. The model endogenizes trading volumes, liquidity supply, information acquisition, and the priority gas auction (PGA) dynamics.

Stage Three (PGA and Trading): Informed traders, possessing private valuations of a risky asset, compete for queue position in a block via a multi-prize all-pay auction (the PGA). Traders bid priority fees to avoid adverse price impact from preceding orders in the block. The model assumes a linear price schedule for the DEX, where execution price impact is proportional to trade size relative to liquidity depth. Traders simultaneously determine their optimal trading volume and priority fee, balancing expected profits against slippage, priority fees, and information costs.
Stage Two (Liquidity Supply): A risk-neutral liquidity supplier sets the DEX's liquidity depth ( $L$ ) to satisfy a zero-profit condition, balancing expected fee revenue from price-sensitive noise traders against expected losses to informed traders (adverse selection).
Stage One (Information Acquisition): A large pool of potential traders decides whether to pay a fixed cost $C$ to acquire private information. The equilibrium number of informed traders is determined by the condition that the marginal trader's expected profit equals the information cost.

The analysis assumes that block time is sufficiently long for substantial price movements to occur, ensuring that private valuations observed at the end of the slot take a definite sign (all buyers or all sellers), thereby simplifying the strategic interaction to a one-sided competition.

Key Contributions and Results

1. Endogenous Selection and the Participation Cutoff
The primary finding is that paid-priority ordering induces an endogenous selection mechanism. In equilibrium, there exists a participation cutoff ( $v_M$ ) for private valuations. Traders with valuations below this threshold find it unprofitable to participate because the adverse price impact from higher-valuation traders (who secure better queue positions) outweighs their potential gains.

Competition Effect: As the number of informed competitors ( $M$ ) increases, the participation cutoff $v_M$ rises. This excludes traders with moderate valuations, leaving only the most aggressive traders (those with valuations near the upper bound $v$ ) to participate.
Price Bias: Consequently, price discovery is hindered. The transaction price becomes systematically biased toward extreme values (the upper bound of the valuation distribution) because the market only reveals information from the truncated, high-valuation tail of the distribution.

2. Liquidity Impairment and Market Shutdown
The selection mechanism has severe consequences for liquidity provision.

Adverse Selection Concentration: Although the fraction of informed traders participating decreases with competition, the trading activity concentrates among the most aggressive participants. This increases the adverse selection cost per unit of liquidity supplied.
Single-Round Clearing: Unlike traditional markets where information is incorporated over multiple short clearing rounds (allowing liquidity suppliers to adjust depth), blockchain markets aggregate all trading into a single discrete round. The liquidity supplier must absorb the full adverse selection risk of the aggregate informed flow in a single decision.
Shutdown Condition: As competition intensifies, the expected losses to the liquidity supplier rise, forcing a reduction in liquidity supply ( $L$ ). The paper derives a viability condition: if the expected losses exceed the fee revenue from noise traders, the equilibrium liquidity depth drops to zero, causing the market to shut down.

3. The Security-Efficiency Trade-off (Block Time)
The paper analyzes the role of block time ( $T$ ), the duration required for consensus.

Dispersion and Truncation: Longer block times increase the dispersion of private valuations (widening the support of the distribution) and reduce the mass of noise trading (due to execution uncertainty).
Amplified Negative Effects: Longer block times raise the participation cutoff further and reduce the effective mass of uninformed demand. This tightens the viability constraints. The authors show that beyond a critical block time, the market collapses because the liquidity supplier cannot cover the adverse selection costs.

4. Empirical Consistency
The model generates stylized facts consistent with empirical data from Ethereum DEXs (e.g., Uniswap v3):

High Transaction Volumes: Per-transaction volumes in blockchain markets are significantly higher than in traditional limit order books, consistent with the model's prediction that only high-valuation (aggressive) traders participate.
Queue Position and Volume: The model predicts that trading volumes decrease with queue position (later in the block). Empirical data confirms this negative correlation.
Priority Fees: The model predicts that priority fees increase with private valuations, which aligns with observed data where informed traders (searchers/bots) pay higher fees.

Significance and Claims
The paper claims that the very features designed to secure blockchain networks—discrete clearing for consensus and paid priority for block builder compensation—fundamentally undermine the viability of these networks as primary venues for price discovery.

Contrast with Traditional Markets: The results stand in sharp contrast to traditional market microstructure theory (e.g., Holden and Subrahmanyam, 1992), where increased competition among informed traders typically accelerates information incorporation and improves market depth. In the blockchain setting, increased competition leads to information truncation, price bias, and liquidity contraction.
Design Implications: The authors conclude that for blockchains to serve as credible infrastructure for price formation, protocol designs must carefully address how transactions are organized and prioritized. The current design, which relies on paid-priority queues and discrete clearing, creates a trade-off where enhancing security (via longer block times) or decentralization may come at the cost of market efficiency and viability.

The paper does not propose specific protocol changes or experimental proposals but rather provides a theoretical warning that the current architectural constraints of blockchains may be incompatible with the requirements of efficient financial markets.