{"paper_id":"2508.09429","title":"Optimal Control of Reserve Asset Portfolios for Pegged Digital Currencies","authors":["Alexander Hammerl","Georg Beyschlag"],"paper_date":"2025-08-13","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2508.09429v1","page_url":"https://thequant.space/flowcharts/optimal-control-of-reserve-asset-portfolios-for-pegged-digit/","scores":{"empirical_rigor":3,"math_complexity":8.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes a stochastic model predictive control framework with a soft‑thresholding reallocation rule and a spread fee, derived via Pontryagin’s Maximum Principle, for managing stablecoin reserve assets.","quote":"We study this treasury problem as a continuous-time control task with two instruments: reallocating reserves between cash and short-duration government bills, and setting a spread fee for either minting or burning the coin. Using Pontryagin's Maximum Principle, we show that the optimal reallocation control exhibits a soft-thresholding structure: no rebalancing occurs when the shadow-cost differential lies within a","summary":"Stablecoins promise par convertibility, yet issuers must balance immediate liquidity against yield on reserves to keep the peg credible. We study this treasury problem as a continuous-time control task with two instruments: reallocating reserves between cash and short-duration government bills, and…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Baseline cash interest rate is zero (rcash = 0).","type":"model","stated":"explicit","quote":"Our baseline assumes rcash = 0, reflecting that vault cash and Federal Reserve master account balances earn no interest, though the framework accommodates rcash̸ = 0 without structural changes.","support":"full"},{"assumption":"T‑bill rate is piecewise constant over each control horizon.","type":"model","stated":"explicit","quote":"(11) For tractability within control horizons [tk, tk+1], we treat the T-bill rate as piecewise constant: rbill(t) ≡r(k) bill for t ∈[tk, tk+1].","support":"full"},{"assumption":"Mint and redemption flows follow exponential Hawkes processes.","type":"model","stated":"explicit","quote":"We use exponential Hawkes processes be-cause they reproduce clustered order flow with a Markovian intensity state, which is what enables the moment-closure forecast and the MPC implementation.","support":"full"},{"assumption":"Fees are set to zero in baseline experiments (δ = 0).","type":"execution","stated":"explicit","quote":"To simplify the interpretation of numerical results, we assume zero mint or redemp-tion fee (δ = 0).","support":"full"},{"assumption":"Linear trading cost is a 1 bp half‑spread and quadratic cost term ρ_ω = 8×10⁻¹³ hour/$.","type":"execution","stated":"explicit","quote":"We set the linear term to a 1bp half-spread, λω = 10−4, so that R λω|ω(t)| dt equals approximately 1bp times traded notional, and ρω = 8×10−13 (in hour/$ units) for the quadratic term.","support":"full"},{"assumption":"Discount rate and hourly conversion of annualized rates are applied as ρ = 0.08/8760.","type":"model","stated":"explicit","quote":"(Over an 8-hour settlement window this corresponds to rbill∆j ≈4.93 × 10−5.) • Discount rate: ρ = 0.08/8760 ≈9.13 × 10−6 hour−1.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Synthetic simulation of mint/burn Hawkes flows and reserve dynamics","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"Experimental Design We conduct numerical experiments over 92 calendar days (August 1 - October 31, 2025) with three equally spaced settlement windows per day (so ∆j = 8 hours), align-ing with major settlement times. To systematically investigate these properties, we define the following market scenarios: (1) Single shock event: The system operates under baseline conditions for days 0-30, experiences elevated","support":"full"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"synthetic"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Linear term 1 bp half‑spread (λ_ω = 10⁻⁴) and quadratic term ρ_ω = 8×10⁻¹³ hour/$ are included in the cost function.","quote":"We set the linear term to a 1bp half-spread, λω = 10−4, so that R λω|ω(t)| dt equals approximately 1bp times traded notional, and ρω = 8×10−13 (in hour/$ units) for the quadratic term.","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Linear term 1 bp half‑spread (λ_ω = 10⁻⁴) and quadratic term ρ_ω = 8×10⁻¹³ hour/$ are included in the cost function.","card_quote":"We set the linear term to a 1bp half-spread, λω = 10−4, so that R λω|ω(t)| dt equals approximately 1bp times traded notional, and ρω = 8×10−13 (in hour/$ units) for the quadratic term.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"The controller preserves most bill carry in calm markets, builds cash quickly under stress, and avoids unnecessary rotations under transient signals.","kind":"performance","quote":"In simulations covering various stress test scenarios, the controller preserves most bill carry in calm markets, builds cash quickly when stress emerges, and avoids unnecessary rotations under transitory signals.","support":"full"},{"finding":"Optimal reallocation exhibits a soft‑thresholding structure with a deadzone set by transaction costs and linear scaling beyond the threshold.","kind":"theoretical","quote":"Using Pontryagin's Maximum Principle, we show that the optimal reallocation control exhibits a soft-thresholding structure: no rebalancing occurs when the shadow-cost differential lies within a deadzone set by transaction costs, and reallocation scales linearly beyond that threshold up to a capacity-imposed saturation limit.","support":"full"},{"finding":"Monotone stress‑response property holds: higher expected outflows or longer windows shift the optimal policy toward cash.","kind":"theoretical","quote":"We also establish a monotone stress-response property: as expected outflows intensify or windows lengthen, the optimal policy shifts predictably toward cash.","support":"full"},{"finding":"In numerical experiments, the window‑based controller outperforms maximum‑yield and maximum‑liquidity heuristics, retaining bill carry while reducing de‑peg risk.","kind":"performance","quote":"The maximum liquidity strategy eliminates depeg risk entirely but forgoes all bill carry, yielding zero revenue across all scenarios. Relative to (i) a maximum-yield rule that de-fers liquidity building until the final settlement window and (ii) a maximum-liquidity rule that sacrifices carry, the controller retains a large share of bill carry in benign conditions while pre-emptively building cash as projected","support":"full"},{"finding":"Maximum‑liquidity strategy eliminates de‑peg risk but yields zero revenue; equal‑weights strategy provides partial improvement but lacks adaptive rebalancing.","kind":"performance","quote":"The maximum liquidity strategy eliminates depeg risk entirely but forgoes all bill carry, yielding zero revenue across all scenarios. The equal weights strategy offers a partial improvement over maximum yield, maintaining a 50/50 target that provides a larger initial cash buffer.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":18,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":2,"extracted_at":"2026-11-14T19:48:36Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}