Paper: arXiv 2606.09003
Authors: Wenpin Tang
Abstract
We consider the interaction between centralized trading and decentralized Proof of Stake (PoS) blockchain ecosystems. Motivated by the increasing dominance of centralized exchanges and the institutionalization of crypto markets, we study how trading activities on centralized exchanges affect staking behavior, token allocation, and decentralization within a PoS blockchain. We formulate a continuous-time mean field model, where the miners simultaneously act as validators in the PoS protocol and traders in a centralized market with price impact. Under suitable assumptions, we establish the local well-posedness of the mean field system, and derive a semi-explicit characterization of the equilibrium trading strategy. Numerical results suggest that centralized trading activities may enhance staking participation, and promote decentralization of the staking distribution through market incentives. We also study the effects of transaction costs and token supply mechanisms on the equilibrium staking ratio and concentration profile. These results illustrate how market microstructure and centralized liquidity provision can exert significant influence on decentralized blockchain protocols.
Complexity vs Empirical Score
- Math Complexity: 8.5/10
- Empirical Rigor: 5.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: This paper presents a highly mathematical mean-field model to analyze the complex interactions between centralized exchanges and PoS blockchains. While the theoretical framework is robust, the empirical validation relies on numerical simulations rather than real-world backtesting, limiting its practical rigor. The novelty lies in applying mean-field games to this specific crypto market microstructure problem.
Research Flowchart
flowchart TD
A["Research Goal: Impact of Centralized Trading on PoS Economy"] --> B{"Key Methodology: Continuous-time Mean Field Model"};
B --> C{"Inputs: Centralized Exchange Dynamics, PoS Protocol Rules, Validator/Trader Strategies"};
C --> D["Computational Processes: Local Well-Posedness, Equilibrium Trading Strategy Derivation, Numerical Simulations"];
D --> E["Key Findings/Outcomes: Enhanced Staking, Decentralization Promotion, Effects of Transaction Costs & Token Supply"];
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper formulates a continuous‑time mean‑field model of miners who act as validators and traders on a centralized exchange, and derives a semi‑explicit characterization of the equilibrium trading strategy.
Source sentence
We formulate a continuous-time mean field model, where the miners simultaneously act as validators in the PoS protocol and traders in a centralized market with price impact. Under suitable assumptions, we establish the local well-posedness of the mean field system, and derive a semi-explicit characterization of the equilibrium trading strategy.
Assumes 6
- Miners are interchangeable, allowing the subscript to be dropped in the representative miner’s objective. model
Source sentence
Without loss of generality, assume that the miners are interchangeable, so we can drop the the subscript ‘k’ in (2.5) to get the objective of a typical miner: U(x) := sup {(ν(t),b(t))} J(ν, b) := E Z T 0 e−βtdc(t) + e−βT [b(T ) + P(T )X(T )] (2.6) subject to X′(t) = ν(t) + N′(t) N(t) −Z(t)X(t), X(0) = x, (C0) dc(t) = rb(t)dt −db(t) −P(t)ν(t)dt −N′(t) L ν(t) N′(t) dt, (C1) b(0) = 0, b(t) ≥0 and 0 ≤X(t) ≤N(t),
- The token supply process N(t) is non‑decreasing and sufficiently smooth. model
Source sentence
Assume that N is nondecreasing.
- Newly issued tokens are automatically staked. model
Source sentence
4We assume that newly issued tokens are automatically staked.
- Market impact is linear in the cumulative order flow with parameter η (Kyle’s Lambda). market
Source sentence
POS, CEX AND MFG 5 where {B(t), t ≥0} is standard Brownian motion, σ > 0 is the volatility, and η > 0 is a market impact parameter (known as Kyle’s Lambda [19]).
Show 2 more
- Transaction cost function L(x) is a power function (e.g., x⁴ or x⁸) in robustness checks. execution
Source sentence
Fixing all the other parameters as above, we consider: L(x) = x4 and L(x) = x8. Figure 3.
- The price dynamics follow P(t)=P(0)+σB(t)−η(Z(t)−Z(0)). model
Source sentence
POS, CEX AND MFG 5 where {B(t), t ≥0} is standard Brownian motion, σ > 0 is the volatility, and η > 0 is a market impact parameter (known as Kyle’s Lambda [19]).
Data
Synthetic data only.
Costs
Modeled
Transaction costs are introduced via power‑law functions L(x)=x⁴ and L(x)=x⁸ in the numerical experiments, affecting the equilibrium trading strategy and staking ratio.
Source sentence
Fixing all the other parameters as above, we consider: L(x) = x4 and L(x) = x8. Figure 3. Figure 3 plots the equilibrium trading strategy, the staking ratio, and the HHI for linear token supply and quartic/octic transaction costs.
Finds 5
- Numerical results suggest that centralized trading activities may enhance staking participation and promote decentralization of the staking distribution through market incentives. theoretical
Source sentence
Numerical results suggest that centralized trading activities may enhance staking participation, and promote decentralization of the staking distribution through market incentives. Numerical experiments demonstrate that centralized trading activities may increase staking participation and promote decentralization of the staking distribution through market incentives.
- The equilibrium trading strategy admits a semi‑explicit form that is independent of the state variable (staked tokens). theoretical
Source sentence
In particular, the miner’s optimal strategy (2.12) is: νeq ∗(t, x) = N′(t)(L′)−1 eβt ePβ,ν(T0)e R T0 t aν(s)ds −ePβ,ν(t) , with ν = νeq, which is independent of the state variable x. That is, the miner’s trading decision is independent of her staked tokens.
- Higher transaction costs (faster‑growing L) suppress trading activity, leading to a slower increase in the staking ratio and a slightly more decentralized staking profile. performance
Source sentence
Comparing with Figure 2, we observe: • Staking ratio increases over time, however, the increase slows as L grows faster (from 0.9 to 0.8 and 0.66 at T0 = 2). This can be explained by the fact that high transaction costs suppress trading activities, and hence, limit the staking ratio. • The staking profile decentralizes over time, and is (slightly) more decentralized as L grows faster (from 6.7 to 6.5 and 6.4 at T0 =
- The staking profile becomes more decentralized over time, especially when token supply grows faster. performance
Source sentence
We observe that the staking ratio increases over time, and the staking profile decentralizes over time and is more decentralized as the token supply grows faster.
- The equilibrium trading rate ν(t) can be either decreasing or increasing depending on the token supply dynamics. theoretical
Source sentence
Also observe that the equilibrate trading rate ν(t) can be either decreasing (Figure 2–3) or increasing (Figure 4).
Checked by others
Supported by
- Reinforcement Learning Pair Trading: A Dynamic Scaling approach
Compare the findings
This paper: Higher transaction costs (faster‑growing L) suppress trading activity, leading to a slower increase in the staking ratio and a slightly more decentralized staking profile.
That paper: Lower transaction costs improve profitability for all techniques, with RL‑based methods trading more frequently when costs are reduced.
Contradicted by
- The Financialization of Proof-of-Stake: Asymptotic Centralization under Exogenous Risk Premiums
Compare the findings
This paper: The staking profile becomes more decentralized over time, especially when token supply grows faster.
That paper: Simulation results confirm rapid phase transition to zero consumer staking under stochastic external market conditions.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.