---
title: HSBC And IBM Show Quantum Can Help Price Bond Quotes – 34% Better Predictions In RFQ Trials
description: HSBC and IBM test hybrid quantum-classical models on European corporate bond RFQs, boosting fill prediction by up to 34% and raising regulatory questions.
author: Darie Nani (Editor-in-Chief)
date: 2025-09-25T14:30:24.000Z
updated: 2026-02-26T18:01:58.180Z
canonical: https://www.sovereignmagazine.com/article/hsbc-and-ibm-show-quantum-can-help-price-bond-quotes-34-better-predictions-in-rfq-trials
categories: Science & Tech
content_type: Spotlight
region: Europe
publication: Sovereign Magazine
---

Loading the [Elevenlabs Text to Speech](https://elevenlabs.io/text-to-speech) AudioNative Player…

HSBC’s quantum-classical experiment produced up to a 34% improvement in predicting whether a corporate bond request-for-quote would be filled at a quoted price, validated on production-scale trading data using [IBM’s Heron processor](https://newsroom.ibm.com/2024-11-13-ibm-launches-its-most-advanced-quantum-computers,-fueling-new-scientific-value-and-progress-towards-quantum-advantage). That number translates into fewer missed opportunities and better automated pricing for the hundreds to thousands of RFQs handled daily in European corporate bond trading.

## How Bond RFQs Actually Work

In over-the-counter corporate bond markets, a request-for-quote represents a customer inquiry asking dealers to bid on a specific trade. Unlike equity markets with central order books, [bond trading](https://www.icmagroup.org/assets/documents/Regulatory/Secondary-markets/ICMA-Secondary-Market-Practices-Committee-European-Secondary-Market-Data-Report-H2-2024-Corporate-Edition-April-2025-020425.pdf) happens between two parties with algorithmic pricing engines making split-second quoting decisions.

Traders face practical constraints: sparse fills, noisy price signals and [uneven liquidity across issues](https://www.sovereignmagazine.com/article/how-nab-s-cloud-revolution-could-transform-australian-trading-technology). [ICMA data](https://www.icmagroup.org/assets/documents/Regulatory/Secondary-markets/ICMA-Secondary-Market-Data-Report-H2-2024-Corporate-Edition-April-2025-020425.pdf) shows GBP-denominated trades accounted for roughly 10% of the €2,567 billion notional traded in European corporate bonds during H2 2024. The scale creates pressure for precision in automated quoting systems.

## What HSBC And IBM Actually Tested

The experiment combined quantum and classical resources, validated on real production-scale European corporate bond data and run on multiple IBM quantum computers including Heron. HSBC used [Qiskit](https://qiskit.org/) to create hybrid workflows rather than replacing entire models.

‘This is a world first in bond trading,’ said Philip Intallura, HSBC Group Head of Quantum Technologies. The improved prediction accuracy for trade fills used IBM’s Heron processor to augment classical workflows and ‘unravel hidden pricing signals in noisy market data’.

Jay Gambetta, Vice President IBM Quantum, explained the collaboration: ‘This exploration shows what becomes possible when deep domain expertise is integrated with algorithm research, and the strengths of classical approaches are combined with the computational possibilities offered by [quantum computing in modern business](https://www.sovereignmagazine.com/article/quantum-computing-in-modern-business).’

## Why Quantum Might Help With Bond Pricing

Quantum circuits can explore high-dimensional optimisation spaces or represent complex probability distributions in ways classical models sometimes struggle with when data are noisy. Hybrid routines can amplify weak signals in feature space that classical algorithms miss.

[IBM’s Heron](https://newsroom.ibm.com/2024-11-13-ibm-launches-its-most-advanced-quantum-computers,-fueling-new-scientific-value-and-progress-towards-quantum-advantage) supports deeper circuits and higher two-qubit gate counts, with the capability to perform up to 5,000 two-qubit gate operations. Current quantum hardware remains noisy and limited in scale, so the reported gains likely depend on careful problem encoding and classical preprocessing. As the [European quantum computing race accelerates](https://www.sovereignmagazine.com/article/european-quantum-computing-race-accelerates-with-rare-five-qubit-system), these developments showcase practical applications beyond theoretical research.

## What We Still Need To Know

Critical methodology questions remain unanswered. HSBC hasn’t specified which classical baselines were used in comparisons — logistic regression, gradient-boosted trees or proprietary models. Dataset details like the number of RFQs, time period, market volatility conditions and train/test splits haven’t been disclosed.

Operational details matter too: how much extra wall-clock time did quantum calls add, and what were cloud usage costs compared with classical compute? Whether results came from live trading or backtests affects their practical significance.

## Market And Business Implications

Better prediction accuracy could materially change quoting behaviour. For a European desk handling 500 RFQs daily, sharper predictions translate into optimised bid-ask spreads and higher executed volumes.

[Other financial institutions](https://multiversecomputing.com/resources/multiverse-computing-study-shows-quantum-algorithm-success-in-financial-services) are funding quantum teams and trials. Ally Financial worked with Multiverse Computing on quantum portfolio optimisation, while academic research shows hybrid quantum-classical approaches gaining traction in finance. However, the industry faces challenges as the [quantum skills gap widens](https://www.sovereignmagazine.com/article/global-quantum-skills-gap-widens-as-industry-races-to-build-future-ready-workforce) across financial services and technology sectors.

Intallura suggested production deployment could follow: ‘We have great confidence we are on the cusp of a new frontier of computing in financial services, rather than something that is far away in the future.’

## Risks And Regulatory Concerns

Potential failure modes include overfitting to historical microstructure, fragility under regime change and operational risk from relying on cloud quantum services. Models mixing classical and quantum steps raise explainability issues for regulators. These concerns mirror broader challenges in [managing concentrated market risk](https://www.sovereignmagazine.com/article/winners-and-losers-a-hedging-strategy-for-concentrated-markets) as technology increasingly drives trading decisions.

Market fairness questions emerge if quantum-assisted pricing becomes a competitive edge. Regulators may scrutinise access asymmetry and auditability of hybrid models, particularly for systematic market-making activities.

If computation becomes a scarce resource in pricing and execution, markets and institutions will need to adapt to a world where specialised processors provide measurable edges. The question becomes how quickly quantum advantages scale and whether they remain accessible to smaller market participants or concentrate among technology leaders.