Bond Market Making with a Hit-Ratio Target
Alexander Barzykin and Axel Ciceri · 2026
Read Bond Market Making with a Hit-Ratio Target by Alexander Barzykin and Axel Ciceri (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-20406.txt.
Introduction
Barzykin and Ciceri study over the counter bond market making with quadratic stock penalty and a size weighted hit ratio target. Dualizing the hit ratio term keeps the reduced Hamilton Jacobi Bellman equation separable, yielding exact optimal controls from inverse fill chances and the Hamiltonian derivative. A quadratic Bergault style approximation gives a Riccati equation for stock curvature, keeping the exact quote map. Linearized, it splits quotes into riskless spread, stock risk correction, and hit ratio correction for multi bond and client tier cases.
Catalog record
| CREATOR | Alexander Barzykin and Axel Ciceri |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38890 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.20406
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bond Market Making with a Hit-Ratio Target Alexander Barzykin and Axel Ciceri Abstract We study OTC bond market making on a size ladder with quadratic inventory penalty and a running target on the dealer’s size-weighted hit ratio within a stochastic optimal control approach. We demonstrate that the corresponding reduced Hamilton-Jacobi-Bellman (HJB) equation remains separable by dualizing the hit ratio target term and provides the exact optimal controls through the inverse of the fill-probability function and the Hamiltonian derivative. We then focus on the quadratic approximation à la Bergault et al. , which yields a Riccati equation for the inventory curvature while retaining the exact quote map. In its linearized form, this approximation produces explicit quote decompositions into riskless spread, inventory-risk correction, and hit-ratio correction. The formulation is general and applies to multi-bond, multi-client-tier scenarios, with special cases obtained by restricting the targeted tiers, their bond coverage, and their associated targets. 1 Introduction Corporate bond trading remains predominantly over-the-counter and dealer-intermediated, even after substantial electronification. In practice, electronification has proceeded less through fully centralized limit-order books than through multi-dealer-to-client request-for-quote (RFQ) protocols, in which a buy-side client solicits a small panel of dealers and typically trades, if at all, with the best respondent. This institutional structure makes bond market making a joint problem of inventory management, counterparty selection, and endogenous execution probability: a dealer must quote competitively enough to win flow, but not so aggressively as to destroy margin or accumulate undesirable inventory. It also means that execution is shaped not only by price, but by the RFQ protocol itself, the number of requested dealers, the response behavior of competitors, and the search technology available to clients. The RFQ mechanism and competition on dealer platforms is an area of active research. Fermanian et al. [ 11 ] provide a foundational empirical model of European corporate-bond RFQs, estimating dealer-quote and client-reservation distributions and explicitly deriving ex-ante hit-ratio functions. Hendershott and Madhavan [ 16 ] analyze the trade-off between search and auction in OTC markets, while Wang [ 23 ] shows that multi-dealer platforms face structural limits in how much price competition they can generate. More recently, Kargar et al. [ 18 ] study sequential search in corporate bonds, and Marín Martínez et al. [ 19 ] develop a causal framework for inference on bond multi-dealer-to-client platforms. Taken together, these papers imply that the dealer’s hit ratio is not an exogenous statistic, but an equilibrium object shaped by platform design, client search, and rival participation. In a related foreign exchange market, Oomen [ 22 ] provides an economic rationale for limiting the number of liquidity providers exposed to each RFQ and for using routing and ranking rules rather than always inviting the full pool. Barzykin [ 3 ] further investigates the role of the win-score promotion gates in aggregator-routed RFQ markets. Recent studies emphasize the importance of inventory, intermediation capacity, and execution quality in bond dealer markets. Bessembinder et al. [ 8 ] survey the fixed-income microstructure literature and stress the centrality of OTC frictions, search costs, and slow-moving electronification. In the corporate-bond setting, Bessembinder et al. [ 7 ] document the importance of dealer capital commitment, Goldstein and Hotchkiss [ 12 ] study dealer behavior in an illiquid market, and O’Hara et al. [ 20 , 21 ] show how execution quality and dealer behavior are altered by electronic RFQ trading. Relationship-based frictions also remain important: Jurkatis et al. [ 17 ] show that bilateral relationships materially affect trading costs in corporate bonds. For quoting models, this body of evidence argues for combining inventory-sensitive pricing with a realistic description of how RFQs translate into fills. One mathematical approach to OTC market making employs the stochastic optimal control framework [ 1 , 13 , 9 , 10 , 14 ] . In particular, Bergault and Guéant [ 4 ] derive general results for OTC market makers with size-dependent client requests and provide dimensionality-reduction techniques suited to large bond universes [ 5 ] . Guéant and Manziuk [ 15 ] address high-dimensional quoting in corporate bonds with a model-based reinforcement-learning approach, while Barzykin et al. [ 2 ] extend dealer-market models to include hedging and market impact, thereby capturing the practical transition between pure internalization and external risk transfer. Bergault and Guéant [ 6 ] further study liquidity dynamics and pricing in RFQ markets using Markov-modulated order-flow models. However, despite the practical salience of hit ratio in bond RFQ workflows, most of this literature treats execution probability in reduced form—typically through quote-dependent intensities or fill probabilities—rather than elevating hit ratio itself to an explicit control target. This omission matters because hit ratio is not merely a post-trade reporting metric. In industry practice, historical hit ratios are used to select dealers and route RFQs, especially for liquid electronic bond orders, alongside axes, response quality, and relationship information. This suggests a natural modelling question: how should a dealer optimally quote when profitability and inventory control must be balanced not just against a passive fill process, but against an explicit target for win probability or hit ratio? The present paper is aimed at this question. Building on the RFQ and OTC market-making literature, we formulate a stochastic control problem in which quote placement is chosen jointly with a penalty for deviations from a desired hit-ratio objective. This preserves the core inventory-risk logic of OTC market making while incorporating a central practical KPI of electronic bond dealing. 2 OTC market making and hit-ratio targeting 2.1 State dynamics and objective We consider a dealer managing a portfolio of bonds offering prices to a group of clients (and/or tiers of clients). Client tiers are indexed by \tau\in\mathcal{T} and bonds by {m\in\mathcal{M}:=\{1,\dots,M\}} . We denote as \mathcal{A}\subseteq\mathcal{T} the subset of client tiers for whom the dealer wishes to target a hit-ratio. Bond m has mid price process dS^{m}_{t}=\sigma^{m}\,dB_{t}^{m}\,, (1) where B_{t} is a standard M -dimensional Brownian motion and the covariance matrix is \Sigma (assumed constant on the risk management timescale). The dealer quotes on both sides of each bond-tier pair across a size ladder indexed by k=1,\dots,K , with deterministic sizes z_{k}>0 . Let s\in\{b,a\} denote side, with b for bid and a for ask. For each bond m , tier \tau , side s , and size k , RFQ opportunities arrive at exogenous intensity \lambda_{m,\tau}^{s,k}\geq 0 . 1 1 1 We require that client tiers have non-trivial total arrival intensity, i.e. \displaystyle\sum_{m\in\mathcal{M}}\displaystyle\sum_{k=1}^{K}\displaystyle\sum_{s\in\{b,a\}}\lambda_{m,\tau}^{s,k}>0 , \forall\tau\in\mathcal{T} . If the dealer quotes offset \delta_{m,\tau}^{s,k}\in\mathbb{R} from mid, the fill probability is f_{m,\tau}^{s,k}(\delta_{m,\tau}^{s,k})\in(0,1)\,, (2) so the OTC fill intensity is \Lambda_{m,\tau}^{s,k}(\delta):=\lambda_{m,\tau}^{s,k}f_{m,\tau}^{s,k}(\delta)\,. (3) Let e_{m} denote the m -th canonical basis vector of \mathbb{R}^{M} . The inventory vector Q_{t}\in\mathbb{R}^{M} and cash process X_{t} evolve as \displaystyle dQ_{t} \displaystyle=\sum_{m\in\mathcal{M}}\sum_{\tau\in\mathcal{T}}\sum_{k=1}^{K}z_{k}e_{m}\,\left(d{N_{m,\tau}^{b,k}}_{t}-d{N_{m,\tau}^{a,k}}_{t}\right), (4) \displaystyle dX_{t} \displaystyle=-\sum_{m,\tau,k}(S^{m}_{t}-{\delta^{b,k}_{m,\tau}}_{t})z_{k}\,d{N_{m,\tau}^{b,k}}_{t}+\sum_{m,\tau,k}(S^{m}_{t}+{\delta^{a,k}_{m,\tau}}_{t})z_{k}\,d{N_{m,\tau}^{a,k}}_{t}\,, (5) where N_{m,\tau}^{s,k} are point processes with intensities \Lambda_{m,\tau}^{s,k}({\delta_{m,\tau}^{s,k}}_{t}) . For any client tier \tau\in\mathcal{T} we can define a notional-arrival scale W_{\tau}:=\sum_{m\in\mathcal{M}}\sum_{s\in\{b,a\}}\sum_{k=1}^{K}z_{k}\lambda_{m,\tau}^{s,k}, (6) and the corresponding size-weighted instantaneous hit ratio r_{\tau}(\bm{\delta}):=\frac{1}{W_{\tau}}\sum_{m\in\mathcal{M}}\sum_{s\in\{b,a\}}\sum_{k=1}^{K}z_{k}\Lambda_{m,\tau}^{s,k}(\delta_{m,\tau}^{s,k}). (7) For the targeted client tiers \tau\in\mathcal{A} , the dealer sets the target levels r_{\tau}^{\star}\in(0,1) . Given running inventory-risk coefficient \phi\geq 0 , terminal inventory penalty coefficient \eta\geq 0 , and hit-ratio weights \kappa_{\tau}\geq 0 , the dealer solves 2 2 2 In practice, the dealer may wish to target subsets