Where trusted data builds confidence

Miao Li runs Debt Capital Markets at Banque Internationale à Luxembourg (BIL) and is a member of the ICMA Luxembourg Regional Committee. In her discussions with David Thomas, Head of Data Trust at LSEG (London Stock Exchange Group), Miao discusses the critical role of data in debt capital markets, explaining how data transforms into actionable insights for client decision-making. The conversation covers data quality, use cases in fixed income markets, challenges, and the importance of data trust and governance.

• Data foundation and narrative: Data must be relevant, complete, accurate, timely, reliable, continuous, comparable, and reusable. It is translated into narratives that clients can understand and act on, combining data expertise with business knowledge.  
• Fixed income market data use: Fixed income markets, which work on individual credit stories, benefit from digitisation starting with structured data collection and moving closer to front-office decisions. Standardised bond market data and issuer information are key areas of development.
• Issuer information challenges: Luxembourg’s market faces difficulties due to many unlisted and unrated issuers, highlighting the need for centralized data repositories to improve pricing and risk models tailored to local markets.  
• Benefits at bond execution level: Automated buy-side processes enhance liquidity support, investor targeting, distribution, and pricing decisions, which is crucial amid market volatility and issuer funding needs.  
• Sustainable finance data needs: Reliable, transparent data on use of proceeds and asset performance is essential for credibility in sustainable and structured finance, despite challenges with qualitative and non-financial information.  
• Operational and risk benefits: Data improves operations, risk management, and control by tracing and analysing situations and anticipating issues, supporting advancements like digital bonds and moving towards faster settlements.  
• Regulatory transparency: Enhanced data reporting equips regulators better and could prevent past scandals by providing real-time transaction data.  
• Data trust challenges: Accountability for data responsibility, consistency across platforms, data quality, and strategic questions about market consolidation and technology investments remain critical. Trusted, well-governed data applied with purpose is necessary for effective data use. 

Miao: Sure, David. Data is the foundation of technology, and I think we have a lot in common across sectors in how we use it. One thing we all agree on is that data-backed arguments are more neutral and often more convincing. But at the same time, we also know we cannot simply send clients all the data we have. That is exactly why we produce so many presentations, charts and visual materials every day. What we really do is translate data into a narrative that clients can understand and act on.

In practice, turning data into meaning starts with data that is relevant, accurate, timely, reliable and consistent over time. But data only becomes truly valuable when it is combined with business judgement and market context, translated into clear insights, and connected across teams, systems and use cases. When those foundations are in place, data does more than inform individual decisions — it helps firms build a more scalable, connected and future-ready operating model.

Miao: Fixed income is often seen as a ‘slow’ asset class compared with equities or FX. While financial markets may share the same macro data, I think the real difference lies on the micro side and in the underlying business processes. While others are trading in the market, in debt capital markets we are working on individual credit stories. When traders on other desks show technical analysis charts, to me it looks like a piece of art :)

But precisely because of that, fixed income has a great deal to gain from better data. Digital transformation began with structured data input and collection from legal and operational perspectives, and then gradually moved closer to front-office decision-making.

So first, it needs to start at instrument level, and data definition and collection are key. Here are two examples that show clearly where things are moving:

• ICMA plays an important role in defining cross-border bond market standards. As market processes become more automated and data-driven, the availability of high-quality, standardised data is becoming a key enabler. A few years ago, it established a designated Working Group, which has recently published the second version of the Bonds Data Taxonomy, including machine-readable definitions of key fields, with examples, expected values and relevant ISO elements.
• ESMA has designed the European Single Access Point (ESAP), which aims to improve access to financial and non-financial information for European issuers. It will start collecting information from national bodies in July 2026.

That brings me to the second level: issuer information, which is a particularly important topic for Luxembourg. As a small economy, many issuers here are not listed or rated, which makes data harder to access. When advising on new issues or debut issuances, we often have to rely on rated companies or data from other countries as reference points. I believe this matters not only for debt capital markets, but also for lending and risk assessment more broadly. If one day we have a centralised data repository for local companies, it would help build pricing and risk models that are better tailored to our domestic market.

That is also why this discussion is particularly relevant for Luxembourg. The market plays a distinctive role in international debt, sustainable finance, and structured finance, yet many issuers in the local ecosystem are not listed or formally rated. Better data availability would therefore not just improve transparency in principle; it could strengthen benchmarking, pricing, risk assessment, and investor confidence in ways that are much more tailored to the realities of the domestic market. 

At the level of bond execution, buy-side processes on platforms are becoming increasingly automated, and trades are being captured and reported more effectively. This creates further benefits for the sell side in three areas:

• Supporting liquidity 
• Identifying target investors and facilitating distribution 
• Strengthening pricing decisions 

That is important, especially in a period of rising volatility, when issuers want to secure funding costs quickly.

Looking ahead, I think the biggest practical gains will continue to come from three areas: better instrument-level data, stronger issuer transparency, and improved execution intelligence. In concrete terms, that means more standardised bond data, better access to company information, and stronger visibility into liquidity, investor targeting and pricing dynamics. These are not abstract improvements — they directly affect how efficiently capital is raised, distributed and managed in the market.

Fourth, data is essential in sustainable and structured finance, where Luxembourg plays a significant role. Use of proceeds and underlying asset performance must be traceable, and impact or risk characteristics need to be evidenced over time. In practice, that is not always straightforward, as much of the relevant information is non-financial and sometimes qualitative. It is crucial to offer reliable and transparent data — both at issuance and on an ongoing basis — because losing credibility in these products would quickly undermine confidence in the asset class as a whole.

Fifth, data significantly enhances operations, risk management, and control functions—not only by tracing, processing, storing and analysing historic and current situations, but increasingly by helping to anticipate and prevent issues before they arise.

This technology-driven efficiency is the backbone of progress in areas such as digital bonds. Even within traditional systems, we are not far from fully transitioning to T+1 settlement in European fixed income markets, and hopefully one day technology will take us to real-time settlement.

For this day to arrive, robust, trusted data will be a critical first step. 

And finally, transparency matters for regulators. With the growing volume of reporting required across the industry, regulators are becoming better equipped. If, for example, real-time transaction data had been available historically, we might not have seen scandals like LIBOR manipulation. 

Miao: I’m far from exhausting all the benefits that data offers us, but it also comes with real challenges. 

First, data trust is fundamentally an accountability issue. We still operate in a system where, when something goes wrong, responsibility must be clearly assigned. Is it the data producer, the data owner, the platform, or the user? Before we reach true “data democracy”, strong due diligence on trusted data sources remains essential.

Second, there are data fragmentation challenges. Are datasets consistent and comparable across platforms and jurisdictions? These points need to be checked before we can compare like with like.

Another challenge is balancing transparency with confidentiality, especially for private issuers. Transparency does not mean disclosing everything; it means making available the information that is relevant, consistent and explainable for the intended audience. That requires trusted frameworks with clear information boundaries, standardised data formats and appropriate access rights, so confidentiality is protected while decision-makers still receive the data they need.

Third comes data quality. Data needs to be curated and quality-checked before use. If technologies like AI are fed with poor or biased data, they don’t correct the problem, they amplify it, creating misleading outcomes. 

This becomes even more important as AI becomes more prevalent. Data and AI can strengthen decision-making, but they do not replace accountability or judgement. We still need to understand data limitations, document assumptions, and test outputs against market reality rather than accept them at face value. AI is most powerful when it improves efficiency and helps identify patterns, but it still needs to operate within a framework of oversight and professional challenge.

And finally, at a strategic level, there are still open questions. Market intermediaries serve clients’ interests in conditions of market inefficiency, and even the best datasets may never be comprehensive enough to capture everything about a bond or an issuer. That is exactly where bankers add value — by matching the interests of issuers and investors, underpinned by due diligence and a deep understanding of both sides. As we are seeing across the financial industry, consolidation continues. Technology investments are long term and costly, which often makes larger institutions the early movers because they can afford to invest even if some initiatives fail. At the same time, technological advances will continue to narrow margins and deepen this process. Will faster consolidation always be good for clients? Or do we risk losing diversity in how services are delivered? I am conscious of time, so let me stop there with that open question.

And this is not only a large-institution issue. Larger firms may be better placed to invest early in technology and data infrastructure, but smaller players often depend even more on trusted external data sources and shared market standards. In that sense, stronger governance and interoperability can help level the playing field, reduce manual work, and make it easier for a broader range of institutions to act with confidence. 

Ultimately, what will determine whether data becomes a true competitive advantage rather than simply more information is trust. Firms need to be able to rely on its quality, understand its origin, apply it consistently, and connect it to real commercial and strategic decisions. Without governance, accountability, and context, more data simply creates more noise. With those foundations in place, however, data becomes a genuine enabler of better client outcomes and stronger market functioning.