Business in Boston: Navigating the market
26 May 2026
The fourth annual 网爆吃瓜 Finance Symposium in Boston saw market participants engage in group discussions on AI, review the current state of readiness for the US Treasury clearing mandate, and share knowledgeable insights on digital assets
Image: Andrew Davis
Opening the fourth annual 网爆吃瓜 Finance Symposium in Boston, emcee Olivia Russell of GLMX welcomed the first panel 鈥� 鈥楽haping the Future of AI in 网爆吃瓜 Finance鈥� 鈥� which explored the shifting conversation of AI within financial services.
The session was moderated by Chelsea Devereaux, head of US asset owners client management, Financing Solutions at State Street, and provided an engaging and interactive discussion among market participants on how they are using AI in their day-to-day.
The speakers identified a number of areas in which AI is used, for example, as a learning assistant or for coding, as well as within client workflows (largely within onboarding workflows) and to achieve a scalable interpretation of legal documents.
Adopting AI models typically starts with a business or technology use case which has defined outcomes 鈥� for example a need for faster client onboarding or easier contract and fee extraction. From there it is a build versus buy decision, according to one speaker. Firms need to determine how AI solutions such as custom copilots, chat bots, and assistance can be embedded into workflows, rather than simply be a stand alone tool.
Commenting on the mention of buy versus build, one panellist highlighted that clients are looking to build for alpha and buy for speed. Firms are more likely to build if the tool is impacting the business and the generation of that. While other firms are looking to partner to help speed up their processes.
Further, it was noted that Copilot, ChatGPT, or other frontier models, are increasingly commoditised. In terms of the hallucination risk 鈥� which occurs when generative AI models produce confident but incorrect, misleading, or fabricated information 鈥� one panellist believes the issue is not as great as it may appear, as these models are becoming increasingly sophisticated and more nuanced in how they answer.
A key point was highlighted in this discussion 鈥� that it is not a model problem, but a data problem. A magnifying glass is being placed on the data, the structure of the data, databases, data lakes, etc. In the end, what will differentiate firms in terms of AI, will be the data and how it is structured.
Shifting the direction of conversion on AI, it was noted that there is some fear and curiosity around what comes next and how the AI journey will evolve. From this, the audience discussed whether there is enough governance around artificial intelligence.
Turning to the audience to share their views, one audience member noted that AI consumes 鈥渟o much energy鈥�. They feared that the largest risk was the loss of independent thinking, adding: 鈥淚f we can鈥檛 contribute to AI and make it even better, AI will dominate humans 鈥� AI will govern humans rather than humans governing AI.鈥�
Another member of the audience noted the importance to not remain complacent in implementation. Despite believing there is a sufficient oversight and governance in where AI is rolled out, the person asked: 鈥淚f there is a one in 1000 chance a model could produce a bad output, if that error later appears, how can firms build a robust system to ensure that output is caught and the risk to the firm is reduced?鈥�
To tackle this issue, panellists noted the need to be transparent and provide traceability, as well as auditability, of all records being produced.
Further, it was stated that 90 per cent of enterprise data is unstructured. Therefore, if a business is using this unstructured data to feed its AI models, the result will be poor. Using a creative analogy to drive home this statement, one panellist said: 鈥淵ou can have a brand new kitchen, new appliances, and a Michelin star chef, but if the food鈥檚 expired, it doesn鈥檛 matter what you鈥檙e cooking, it is still going to taste bad.鈥�
Concluding the discussion, market participants were reminded that training is important. They were advised to be critical of answers that come from AI models when asking a broad question, and that they should also refine their prompts and add files where possible.
Challenges remain for firms facing US Treasury clearing mandate
Market participants reviewed the next steps required for the implementation of the mandatory US Treasury clearing mandate. With no shortage of choice, the panel reviewed the current central clearing models on offer 鈥� Sponsored, Agent Clearing Service (ACS), Collateral-in-Lieu 鈥� and which is best for which clients.
Providing a comprehensive breakdown, Sofia Pavlidou, product lead for central clearing services on BNY鈥檚 Global Collateral platform, noted that there is no one-size-fits-all solution in regard to the central clearing services being launched, but there is optionality in the market.
She stated: 鈥淪ponsored GC was launched a couple of years ago, and since the finalisation of the mandate at the end of 2023, it has increased more than 400 per cent, crossing over US$800 billion in daily volumes. This shows that the market is moving into cleared channels ahead of the mandate.鈥�
Jeff Sowell, head of Financing Solutions product strategy, North America at State Street, commented that the Sponsored model was launched by FICC in 2005 and, up until 2017, was 鈥渁 really niche鈥� model.
鈥淲hen that model was created in 2005, I don鈥檛 think anyone in their wildest dreams could have imagined that it would turn into the US$2.4 trillion wagon that it is today,鈥� he added.
He referred to the legacy Sponsored model of 2005 as the 鈥渏ack of all, master of none鈥�. A model which worked well through challenging market conditions, but one that does not offer the same efficiency as the new models which better fit specific client archetypes.
Collateral-in-Lieu is an extension of the Sponsored general collateral (GC) service, explained Pavlidou. The Fixed Income Clearing Corporation (FICC) takes a lien on the assets that sit on the cash investors鈥� triparty account, and because of this lien, the model is able to offer margin and capital benefits if clients are risk-weighted asset (RWA)-constrained.
This is possible because, in most circumstances, there is no need to collect margin on behalf of the Sponsored member, and there is no need to guarantee the performance of the trade, which can reduce RWA exposures.
Sowell added that this model is well-suited for institutional cash investors, who have a focus on yield, liquidity, and are more sensitive to counterparty credit risk. The model also provides the operational ease of having a triparty agent bank hold the collateral and that undertakes the allocation, sending, reporting etc.
The ACS is the next model BNY has launched and is an extension of the existing Agent Clearing Service (the bilateral version) that the FICC offers, except this new model is on triparty. Essentially, this model enables members to continue to receive the benefits of the service by leveraging the operational benefits of operating under a triparty structure, as well as the net margin benefits that the ACS model provides.
For Sowell, the ACS model is well suited for firms like alternative investment funds, which are focused on financing treasuries and minimising all-in cost as it is the most efficient buy side access model that FICC offers.
Looking forward, Sowell predicts: 鈥淲e鈥檙e going to see new balances come directly into the new models, and then existing activity will migrate out of Sponsored into those new models over time.鈥�
Moving onto a solution which has created much discussion, Pavlidou discussed Done-Away. She said: 鈥淔rom a Global Collateral standpoint, when we think about Done-Away, it is important to have the operational structure to support it, which we have built on our triparty infrastructure. As we get closer to the mandate, dealers might realise that they don鈥檛 have as much capacity to either clear or paper more counterparties, this is where Done-Away will be a solution that is very beneficial in helping the market comply with the mandate.鈥�
Readiness
The conversation shifted to discuss the current state of readiness of firms in both the US and Europe, with moderator Andrew Lazar, managing director, head of rates sales at Buckler 网爆吃瓜, asking: where do we find ourselves right now?
Alice Elizabeth Othigo, director, global product manager at BNP Paribas, noted: 鈥淭his regulation is a major shift for the industry and for the current market, and the industry is preparing for this change. Yet the preparatory steps and the readiness of different firms may vary depending on market segment, regional presence, clearing model options, and project headways.鈥�
In terms of the fundamentals that firms are working on in order to be ready for this regulation, Othigo notes a number of important factors. First, assessing trading patterns, identifying eligible transactions, engaging with trading counterparties and clearing houses to define the best-suited clearing model. Followed by legal repapering and streamlining the flow to avoid activity bottlenecks and drawbacks.
Then, running a comprehensive programme to adapt technology and operating models to the new environment and latest workflows features. As a result, delivering scalable and integrated service models supporting firms鈥� and their clients鈥� business strategies and goals.
For Andy Hill, managing director, co-head of Market Practice and Regulatory Policy at the International Capital Market Association (ICMA), it is as much about awareness as preparedness.
US Treasuries are traded globally, and for ICMA, the association has members in 70 different jurisdictions, all of which trade US Treasuries to some extent or another. It is no longer simply about whether these jurisdictions are getting ready, but whether they know they are in scope.
In terms of non-US members who are fully prepared, an ICMA survey from late 2025 reports that only four per cent are fully prepared, while around half have not begun or are just starting. As a result, the association is working to raise awareness.
The survey also highlighted a number of current challenges facing in-scope firms, the two key points remain around understanding the cross-border application and whether firms are in-scope. Other issues include coordination with counterparties, regulatory compliance and documentation, operational readiness, and choosing access models.
Notably, Hill highlighted a potential negative impact of the mandatory clearing mandate. He explained that from a non-US entity perspective, if US Treasury and Treasury repo is not part of a firm鈥檚 core business, then firms must ask 鈥� considering the operational lift, the documentational lift, and even the regulatory uncertainty of being in scope 鈥� if they need to use US Treasuries or if they can use something else.
As a result, he anticipates that a number of marginal users of the Treasury market will likely step out, due to the complication and expense of this regulatory move.
Further, Hill said there remains uncertainty around triparty in a European context. He explained that if a firm moves forward with a triparty which starts with US Treasuries as part of the basket, they are in scope. But if they start off without US Treasuries and these are later caught up in that triparty due to optimisation engines and dynamic collateral allocation, the regulatory requirement remains unclear.
In his conclusion, Hill stated that there is still much to work through from a European perspective, and in a global context in general.
It鈥檚 all about data
The Data Quality and Controls and Oversight in Regulatory Reporting panel explored how data is used in securities finance, how it impacts the decisions made by industry participants, and how it looks at risk management and compliance.
Gavin Marcus, head of North American sales at S&P Global Market Intelligence Cappitech, began the panel by explaining how the firm is evaluating data quality, highlighting that regulation is the key driver. He noted that feedback from clients, regulators, and industry bodies, suggests that firms are looking at data quality now as they cannot 鈥済et the house in order鈥�. He underscored the importance of analysing their data going back to source systems such as trade capture, how firms are booking the trades, risk systems, order management systems, and if the reference data is properly established.
Building on Marcus鈥� remarks, Thomas Veneziano, director of product management at Broadridge, said that data quality is now driving much of the industry鈥檚 decision making. He noted that with greater clarity around regulatory timelines, market participants have the opportunity to improve data quality, increase transparency, and engage more proactively with regulators on how the date should be used. Previously, tight timelines meant firms were focused primarily on meeting reporting requirements rather than ensuring quality data.
鈥淒ata is no longer a back-office or reporting issue or conversation,鈥� according to Tommy Ros, senior vice president, capital markets services at Delta Capita. 鈥淭here is operational friction that now moves into pricing, equity, counterparty confidence, regulatory standing, and reputational damage.鈥�
He noted that the lack of regulatory change has led to an emphasis across firms looking at the change in speed and scale of their current operating models due to data flow from front-to-middle-to-back office now being almost instant, causing traditional, manual controls to feel outdated, and raises the question of how can firms improve their data quality as the regulator is putting such emphasis on it.
Concluding this section of the panel, Marcus noted that he has had numerous conversations with clients that suggest that data lakes are 鈥渋mperative鈥�, with firms analysing their data across the front, middle, and back offices as well as for regulation. 鈥淔irms have put the data lakes in place so that all the data can be governed correctly, with all the data points and elements needed for processing, pricing, liquidity, and regulation.鈥�
Panel moderator Nancy Steiker, senior director, Global 网爆吃瓜 Finance Product Management at FIS, then posed the question: why does data quality matter?
鈥淨uality is extremely important,鈥� said Veneziano, noting this is true for a number of reasons. Firms rely on a wide range of analytics to assess counterparties and determine appropriate trading strategies. He added that, from a client onboarding perspective, several variables must be considered, including standing settlement instructions (SSIs), concentration limits, and proper investment guidelines.
鈥淗aving data in a proper, verified, clean, quality format is key, because if a firm experiences a fail today, it may face TMPG claims, CSDR penalties, reputational risk, and potential buy-ins,鈥� he said, explaining that data impacts reputational risk, market risk, and financial risk.
Ros emphasised the importance of data quality, saying that regulators are now requesting to know where data comes from, who owns it, and how it can be trusted. 鈥淭raceability is a huge point of emphasis today, which opens up a number of questions when looking at third party and AI-integrated models across third party regulatory platforms or in-house builds.鈥�
Shifting the conversation to AI, Marcus noted that 鈥渆very single client鈥� is asking how AI is being used in tools and how AI can be used to create efficiencies and cost savings. He said at S&P, AI is being embedded in all S&P solutions as well as being used internally in everyday processes, with everybody being encouraged to evaluate ways they can use agents or tools to embed AI in their processes while keeping a human 鈥榠n the loop鈥�.
He adds the caveat that guardrails are needed, that code cannot be installed on desktops, that it needs to be vetted by the IT and security teams. 鈥淭he last thing you want is to see client information on the public domain because someone has made a mistake,鈥� he explained, highlighting that AI implementation has to be done in a slow, controlled manner to ensure it is done correctly.
The panel then moved on to discuss governance, with Veneziano underscoring that governance is shared across the organisation, and that middle office operations play a key monitoring role. He added that governance requirements vary depending on the data type, but also noted that operational risk, trading and market risk, compliance, and internal auditors all play a role in reviewing data as well as regulators.
Marcus added that data scientists are now being employed as data stewards across the industry.
Commenting on best practices to ensure front, middle, and back offices work together, speaking from Delta Capita鈥檚 perspective, Ros highlighted that building an ecosystem across the post-trade world has resonated with their client base in the past, having the ability to integrate with various systems that clients are using, and streamlining it into one platform that can aggregate, normalise, and present the data in an easily-digestible format for users across the board.
Moving onto the risks associated with poor-quality data and fragmented data management Veneziano noted that both financial and market risks can arise. He explained that if data is inaccurate or poorly maintained, risks such as incorrect SSIs may occur. He stressed the importance of maintaining strong controls to ensure that settlement instructions remain accurate. Poor data, he added, can create issues across both new and existing transactions, leading to potential fines or penalties.
Ros added that, as the industry becomes faster and the flow becomes almost instantaneous, front-to-back office firms no longer have the luxury to fix downstream regarding errors they have made in trade transaction reporting, adding another layer of risk across the board.
The importance of data transparency was echoed by every panellist, particularly the ability to trace the lifecycle and identify where (if at all) an error occurred. From a reconciliation perspective, Veneziano said firms need to ensure that data is accurate, properly maintained, and sourced from reliable systems.
鈥淭racebility has moved from being a nice-to-have to an expectation in today鈥檚 world, especially in legacy and fragmented environments that many firms still have, but the regulators do not care about complexity. The feedback we have received from participants is that having a golden trail of data lineage and traceability is becoming a huge point of emphasis,鈥� Ros added.
Concluding the discussion, market participants heard that data is the new goal, with firms wanting to do less with more 鈥� data is the way for firms to differentiate themselves, create new processes, be more efficient, and increase revenues.
The agent lender perspective
Agent lending is only commoditised if participants run a generic, undifferentiated programme, with market leaders distinguishing themselves through customisation, collateral and capital optimisation, and smart use of technology and data to align closely with clients鈥� evolving financing and regulatory needs, according to panellists.
Titled: 鈥楤eyond the baseline: how agent lenders compete in a commoditised market鈥�, participants reviewed how firms leverage collateral optimisation, proactive regulatory guidance, balance sheet and risk-weighted asset (RWA) management expertise, and which service innovations resonate with beneficial owners and where the market is heading as competition intensifies.
Jon Whiting, vice president, head of international and fixed income agency trading at Fidelity Investments, laid the foundation for the panel through his perspective on differentiated models from a beneficial owner standpoint.
鈥淲hat we are trying to do is ensure that we are extrapolating as much value as possible from the parameters our clients provide, including risk parameters, programme parameters, and cultural parameters.鈥�
He emphasised the need to maintain stability to ensure the firm can make as many transactions as possible in addition to ensuring there is a level of connectivity with street-side firms so they know the challenges and can provide mutually beneficial solutions.
鈥淔rom Fidelity鈥檚 Agency Lending鈥檚 perspective, we are trying to ensure that we are building a sustainably scalable model. This means we are keeping an eye on costs and maintaining efficiency while being a driver of technology solutions for our client base and the wider market.鈥�
Leading the panel forward, moderator Brian Morrissey, executive director, Americas buy side trading services sales, securities services, at J.P. Morgan, asked the panel what collateral optimisation means to an agent lender, and from a beneficial owner perspective.
Mark MacNeill, head of US equity and corporate bond trading at eSecLending, answered, highlighting that as the market is growing and becoming commoditised, a lot of performance at the margin can be generated from managing collateral well.
MacNeill states that managing collateral well, particularly through accurate cash forecasting and efficient intraday decision-making, or even smaller improvements, such as timing cash investments and repos more effectively, can add 10鈥�15 basis points, which becomes significant when applied to large balances.
He also underscored that by understanding the needs of counterparties, firms can optimise collateral deployment and create more value on the lending side.
Alvin Oh, global head of securities lending product at Wematch.live, adds that, from a trading platform perspective, there is a need to introduce effective utilisation of inventory, and the question of 鈥榟ow do we allow agent lenders to earn more and lend more effectively on platforms such as Wematch鈥� needs to be answered. 鈥淚t is through analytics, workflow automation, and smarter matching tools that we can help firms maximise their lendable assets, reduce inventory, rather than lending up and having to finance it overnight, and bring about collateral that they can then run their models against,鈥� Oh concluded.
Moving on to answer the question of what clients are asking of agency lending programmes and discussing regulatory impacts, Whiting notes that clients are seeking preparedness for structural market changes such as T+1, Treasury clearing, and equities as collateral in the US 鈥� 鈥渃lients want to know what is going to impact their programme and their day-to-day as it currently stands鈥�.
Anthony Toscano, North American head of global securities lending solutions at MUFG Investor Services, shifts the discussion to the ETF market, highlighting that the firm serves a diverse client base, including large institutional investors and quickly growing ETF providers 鈥� many clients are seeking help on central clearing as simple as the buys and sells of their treasury positions.
Clients are also constantly wanting to revisit their bespoke programmes, and are looking to see what the implications of tweaking their current guidelines would be 鈥� whether broadening their collateral sets or changing buffers or the dynamic liquidity tests there are prior to putting a security on loan.
鈥淓TFs in particular, continue to grow rapidly, meaning the buffers that have been used may be slightly too conservative. Being a full service agent lender means, not only are we finding distribution channels for this supply, being adaptable to the collateral types to address the binding constraints of the counterparts, but also looking at the client guidelines and parameters to see if there鈥檚 things that can be tweaked,鈥� said Toscano.
From a J.P. Morgan perspective, Morrissey notes that clients are continuously asking that the agent lender sits more within the financing ecosystem, and they are looking for providers that are able to handle the full lifecycle of financing, such as liquidity management, collateral optimisation, and seamless connectivity to financing and margin workflows and revenue generation.
Furthering the discussion into service implementations, Oh comments on the use of a trading venue and electronic marketplace that allowed the firm to open up broader liquidity pools rather than relying on traditional bilateral relationships. He noted that this means clients can access a wider network of counterparties, therefore creating more opportunities for supply and demand and improving overall market efficiency in addition to providing market transparency.
Concluding the panel, MacNeill underlines the importance of agent lenders to consider the opportunity cost of a loan versus cash, or a loan versus non-cash. He also notes that for clients that primarily might be in overnight vehicles that can take equity collateral, this provides a good opportunity and may allow them to perform better from an intrinsic standpoint, and that is important to make sure clients see incremental performance.
Tokenisation moves from theory to infrastructure
The 鈥楤lockchain, Digital Assets, and the Tokenisation of 网爆吃瓜 Finance鈥� panel brought together perspectives from across the securities finance ecosystem, with participants largely aligned on one point: digital assets are no longer a theoretical discussion.
Moderated by Tina Joshi, head of securities finance and collateral management solutions at Broadridge, the panel explored how institutional thinking around tokenisation has evolved, what is still holding adoption back, and whether the industry is genuinely approaching a large-scale inflection point.
Opening the discussion, Joshi asked whether the industry is now at an inflection point or still firmly in the infrastructure build-out phase. Sasha Sitsker, vice president, solutions engineering at EquiLend, argued that the answer is both. 鈥淭here was a lot of apprehension about digital assets,鈥� he explained. 鈥淣o one knew exactly what stepping into the digital asset space represented, or what the benefits necessarily were.鈥�
According to Sitsker, the industry has now moved beyond that uncertainty, with early adopters and major market participants beginning to align around what future infrastructure may look like and the tangible benefits of its adoption. 鈥淎t least now we know which infrastructure we are building out to,鈥� he noted.
Nathaniel Lindsay, vice president and head of electronic trading, agency securities lending at State Street, agreed, but highlighted the importance of the evolving US regulatory landscape in accelerating institutional interest.
He pointed to the January 2025 executive order supporting digital assets in the US, alongside the US 网爆吃瓜 and Exchange Commission鈥檚 (SEC鈥檚) rescission of Staff Accounting Bulletin (SAB) 121 through SAB 122.
Lindsay explained that SAB 121 effectively prevented prudentially regulated institutions from engaging meaningfully in digital asset custody because it required assets under custody to sit on balance sheet. 鈥淚magine if State Street鈥檚 US$50 trillion was on balance sheet,鈥� he remarked. 鈥淚t is completely infeasible from a regulatory capital standpoint.鈥�
Steve Everett, chief commercial officer CDS and head of post-trade innovation at TMX Group, meanwhile argued that the infrastructure discussion needs to be viewed more holistically. While many digital asset discussions focus on innovation layers and settlement rails, Everett suggested the industry often overlooks the operational bottlenecks inside firms themselves. 鈥淣o matter how efficient your rails are, when a client back-end infrastructure does not quite meet up to the same speed, it all goes down to the lowest common denominator,鈥� he observed.
When asked about which part of the industry could see scale adoption of digital assets, Everett highlighted private markets as one area where digital transformation may accelerate more quickly because, unlike many public markets, there is little entrenched infrastructure to replace. 鈥淧rivate markets actually have very little infrastructure, certainly in Canada and it鈥檚 mostly paper based,鈥� he said.
A major theme throughout the panel was the growing distinction between cryptocurrency and broader digital asset infrastructure.
Sitsker noted that for many years digital assets were effectively treated as synonymous with cryptocurrencies such as Bitcoin and Ethereum. Now, however, the industry is developing a much more nuanced understanding of the space, and is better able to distinguish between the asset class and the underlying technology. This includes tokenised securities, tokenised real-world assets, stablecoins, and even the use of distributed ledger technology (DLT) purely for operational workflows.
As an example, Sitsker referenced EquiLend鈥檚 1Source distributed ledger platform, which creates a single onchain version of a securities lending contract accessible to both counterparties. 鈥淚f adopted at scale, it completely eliminates the notion of counterparties being out of sync or needing to come and reconcile their contracts on a day-to-day basis,鈥� he said.
Lindsay added that State Street鈥檚 strategic approach to digital assets has also evolved significantly. 鈥淥ur 2026 gross technology investment in digital has grown orders of magnitude since 2025 and even more substantially since 2024,鈥� he revealed.
According to Lindsay, digital assets have shifted from being a niche custody discussion to something being assessed strategically across the entire organisation.
The question of complexity versus simplification was another central topic during the discussion.
Asked whether tokenisation simplifies custody operations or makes them more complex, Lindsay acknowledged that the short-term reality is likely more difficult operationally.
Maintaining synchronisation between traditional securities and their digital representations, he explained, introduces challenges around minting, burning, corporate actions, stock splits, and dividend processing. However, he argued that in a fully digitally native environment, complexity could ultimately decrease significantly. 鈥淭hat is where we start to reap some of the more compelling benefits of this technology, like programmability and atomic settlement,鈥� he noted.
On the question of what still prevents institutions from moving faster into digital assets, the panel broadly agreed that the primary challenge has shifted away from regulation and towards technology integration.
Sitsker argued that the industry now broadly understands the regulatory framework, and the focus has instead turned towards scaling adoption and reducing integration pain points. 鈥淩ight now everyone has got their sights focused on: okay, we know what the rules of the playground are,鈥� he said.
Lindsay similarly highlighted the importance of education and institutional-grade security. He explained that many clients still associate digital assets primarily with cryptocurrencies, rather than with tokenised representations of traditional financial instruments. 鈥淲e think about abstracting away complexity and really being a bridge between traditional and digital markets for our clients,鈥� Lindsay said.
The concept of institutional-grade infrastructure became a recurring theme throughout the discussion. From a vendor perspective, Sitsker argued that many of the industry鈥檚 early assumptions around decentralisation are now being reconsidered.
Rather than prioritising decentralisation, he suggested institutional markets instead require scalability, security, and trusted operators. 鈥淭he decentralisation angle was probably more of a hindrance than a benefit to an industry that needs clarity around who is maintaining these networks and applications,鈥� he explained.
Everett agreed, referencing Canada鈥檚 Project Samara as an example where assumptions around disintermediation did not ultimately hold up under scrutiny. 鈥淭he starting point should not be disintermediation,鈥� he argued. 鈥淭he starting point should be: what value is being derived and what exactly is the problem we are trying to solve for?鈥�
The panel also challenged some of the industry鈥檚 assumptions around atomic settlement.
While often presented as one of tokenisation鈥檚 key advantages, Everett questioned whether marginal settlement speed improvements alone justify the operational transformation required.
鈥淚s atomic settlement as big a value as we say it is?鈥� he asked. 鈥淭oday settlement in traditional infrastructure is measured in 10ths of second. The issue isn鈥檛 the technology, it鈥檚 more about how the ecosystem connects with each other today and the associated behaviour which is concomitant.鈥�
Instead, Everett highlighted programmability, reconciliation reduction, and cross-border collateral mobility as potentially more meaningful long-term benefits as opposed to pure atomic settlement as a value driver. 鈥淐ross-border mobilisation of securities is really not great,鈥� he said, suggesting digital infrastructure could materially improve that process.
The discussion later turned towards competition and market structure.
While the panellists generally agreed that trusted incumbents are likely to remain dominant in institutional markets, they also acknowledged that barriers to entry are falling rapidly.
Everett suggested advances in technology and artificial intelligence are making it significantly easier for new firms to build digital infrastructure. 鈥淭he barrier to entry dropping means that you are going to naturally get more competition,鈥� he said.
Still, both Lindsay and Sitsker argued that network effects and institutional trust will continue to favour large incumbent providers, particularly in custody and asset servicing.
Finally, Joshi asked the panel which asset classes they believe will first achieve genuine tokenised scale beyond pilot programmes.
Sitsker pointed towards stablecoins, describing them as the most easily understood digital asset use case and potentially foundational infrastructure for digitally native transactions.
Lindsay instead highlighted tokenised money market funds, citing growing institutional demand for yield-bearing digital cash instruments that can also be mobilised as collateral. 鈥淭he product-market fit is pretty clear,鈥� he said.
Everett agreed with the money market fund thesis as collateral, but also argued that private markets may ultimately see the most complete digital migration due to their lack of entrenched infrastructure. 鈥淲hen that (private markets) moves across to this technology,鈥� he concluded, 鈥淚 think it is going to move at scale.鈥�
Away from the silo
The 鈥楥ollateral Management and Optimisation: Making the Most of What You Have鈥� panel highlighted one clear message: collateral optimisation is no longer just a back office efficiency exercise.
Moderated by Amy Caruso, head of collateral initiatives at ISDA, the discussion explored how firms are trying to put the right collateral in the right place at the right time, while navigating constraints around capital, return on equity, liquidity, and operational complexity.
Opening the discussion, Caruso noted that collateral optimisation can mean different things to different firms. Some focus on capital, others on return on investment, and others on ease of movement.
For Joseph Pirro, director and head of equity lending and repo Americas at ING, the answer has been a major internal overhaul.
Pirro explained that ING鈥檚 securities finance division had historically been split across fixed income repo, equity lending and equity repo, and linear equity derivatives, with each area managing its own funding and collateral.
That siloed model has since shifted towards a more centralised approach through a Funding and Liability structure at ING, which has now been in place for many years with Global 网爆吃瓜 Finance. The project is a bank-wide initiative to make Global 网爆吃瓜 Finance the hub for all collateral within the bank.
鈥淲e are bringing it all back and doing it all from an optimisation perspective, but it all starts with the data and making sure that it is clean,鈥� Pirro said.
He added that the project is still in its early stages and is expected to take three to five years, with a focus on dashboards, system connectivity, front-end booking systems, and reducing manual intervention.
Ted Leveroni, head of margin services at BNY, said buy side firms face a similar problem, even if the details differ. 鈥淭he problem statement, I think, with buy side and sell side, is universal and similar,鈥� he said. 鈥淚t is inefficient operations leading to inefficient use of assets, leading to a drag on performance.鈥�
For Leveroni, the first step is an aggregated view of assets, followed by optimisation tools and mobilisation capabilities. He said firms need to see not only their long assets, but also collateral received through financing and derivatives activity.
Ed Corral, head of collateral services at Pirum, argued that vendors can help firms accelerate this process by reducing internal friction and supporting consolidation of inventory and eligibility schedules.
Reflecting on his own experience, Corral said optimisation can take years, but the value can be substantial. 鈥淲hen you finally get there, the value is incredible,鈥� he said.
From a central counterparty perspective, Harsh Devpura, director, securities finance at OCC, highlighted the role of infrastructure modernisation.
Devpura pointed to OCC鈥檚 Ovation programme, describing it as an investment to comprehensively redevelop and modernise the company鈥檚 risk management, clearing, and data systems.
According to Devpura, the platform is moving from a single mainframe to microservices in the cloud that can support faster product delivery, optimise workflows, provide real-time data that can act as tools for members to develop their own applications, and enhanced security and improved resiliency.
He also highlighted OCC鈥檚 exploration of new membership access models and tri-party services.
The panel then turned to artificial intelligence, with speakers broadly agreeing that the technology could have a significant role in collateral optimisation.
Corral continued that optimisation is particularly well suited to AI because many of its objectives can be represented mathematically. 鈥淓very complex objective in optimisation can be represented as a formula, so it is tailor-made for AI,鈥� he said.
Pirro said ING is already using AI for admin and to help with on the desk coding projects. He identifies possible future use cases the market could consider, such as digitising agreements, reviewing collateral support annexes, and summarising regulation.
Leveroni suggested that AI could ultimately support pre-trade optimisation, helping clients identify the best source of financing for a particular security at a particular time. However, he cautioned that firms still need the underlying data, connectivity, and asset mobilisation capabilities before they can fully benefit.
Devpura added that firms can use analytics to assess what their margin requirements may look like before portfolio finalisation.
The use of non-cash collateral was another major theme.
Leveroni said the driver behind greater use of securities as collateral is simple: 鈥淚t is cash drag.鈥� He argued that most collateral support annexes remain cash-only, creating a need for standardisation, automation, and more efficient ways to update documentation at scale.
Pirro agreed, noting that equities and Treasuries have long been used in securities lending, but that large pools of buy side assets remain idle. 鈥淭he idea of being able to mobilise a lot of these things that were idle, even if you just mobilise it another step, is a step in the right direction,鈥� he said.
Corral recalled previous efforts to post non-cash collateral under existing agreements, only to find that counterparties were not operationally ready to accept it. 鈥淭here was nothing more to it than that,鈥� he said, recalling one operations manager鈥檚 explanation that cash was simply easier to post.
Finally, the panel considered tokenisation and its potential impact on collateral management.
Leveroni argued that tokenised collateral could support intraday margin calls and ultimately better pricing. 鈥淭o me, you are taking an inefficiency out of the market, which translates into cost out of the market, which is better for the investor,鈥� he said.
Corral highlighted the value of models that link traditional collateral with tokenised representations, suggesting these hybrid structures may be more realistic in the near term than a fully digital replacement of existing markets.
Devpura said tokenisation could also support extended trading hours, particularly in a 23/5 or 24/7 market environment. 鈥淚f you really want to get into a full 24/7 or 23/5, tokenisation becomes a strong workflow tool,鈥� he said.
Across the discussion, the message was consistent: collateral optimisation requires technology, but it begins with visibility, clean data, and the ability to mobilise assets across fragmented systems.
The session was moderated by Chelsea Devereaux, head of US asset owners client management, Financing Solutions at State Street, and provided an engaging and interactive discussion among market participants on how they are using AI in their day-to-day.
The speakers identified a number of areas in which AI is used, for example, as a learning assistant or for coding, as well as within client workflows (largely within onboarding workflows) and to achieve a scalable interpretation of legal documents.
Adopting AI models typically starts with a business or technology use case which has defined outcomes 鈥� for example a need for faster client onboarding or easier contract and fee extraction. From there it is a build versus buy decision, according to one speaker. Firms need to determine how AI solutions such as custom copilots, chat bots, and assistance can be embedded into workflows, rather than simply be a stand alone tool.
Commenting on the mention of buy versus build, one panellist highlighted that clients are looking to build for alpha and buy for speed. Firms are more likely to build if the tool is impacting the business and the generation of that. While other firms are looking to partner to help speed up their processes.
Further, it was noted that Copilot, ChatGPT, or other frontier models, are increasingly commoditised. In terms of the hallucination risk 鈥� which occurs when generative AI models produce confident but incorrect, misleading, or fabricated information 鈥� one panellist believes the issue is not as great as it may appear, as these models are becoming increasingly sophisticated and more nuanced in how they answer.
A key point was highlighted in this discussion 鈥� that it is not a model problem, but a data problem. A magnifying glass is being placed on the data, the structure of the data, databases, data lakes, etc. In the end, what will differentiate firms in terms of AI, will be the data and how it is structured.
Shifting the direction of conversion on AI, it was noted that there is some fear and curiosity around what comes next and how the AI journey will evolve. From this, the audience discussed whether there is enough governance around artificial intelligence.
Turning to the audience to share their views, one audience member noted that AI consumes 鈥渟o much energy鈥�. They feared that the largest risk was the loss of independent thinking, adding: 鈥淚f we can鈥檛 contribute to AI and make it even better, AI will dominate humans 鈥� AI will govern humans rather than humans governing AI.鈥�
Another member of the audience noted the importance to not remain complacent in implementation. Despite believing there is a sufficient oversight and governance in where AI is rolled out, the person asked: 鈥淚f there is a one in 1000 chance a model could produce a bad output, if that error later appears, how can firms build a robust system to ensure that output is caught and the risk to the firm is reduced?鈥�
To tackle this issue, panellists noted the need to be transparent and provide traceability, as well as auditability, of all records being produced.
Further, it was stated that 90 per cent of enterprise data is unstructured. Therefore, if a business is using this unstructured data to feed its AI models, the result will be poor. Using a creative analogy to drive home this statement, one panellist said: 鈥淵ou can have a brand new kitchen, new appliances, and a Michelin star chef, but if the food鈥檚 expired, it doesn鈥檛 matter what you鈥檙e cooking, it is still going to taste bad.鈥�
Concluding the discussion, market participants were reminded that training is important. They were advised to be critical of answers that come from AI models when asking a broad question, and that they should also refine their prompts and add files where possible.
Challenges remain for firms facing US Treasury clearing mandate
Market participants reviewed the next steps required for the implementation of the mandatory US Treasury clearing mandate. With no shortage of choice, the panel reviewed the current central clearing models on offer 鈥� Sponsored, Agent Clearing Service (ACS), Collateral-in-Lieu 鈥� and which is best for which clients.
Providing a comprehensive breakdown, Sofia Pavlidou, product lead for central clearing services on BNY鈥檚 Global Collateral platform, noted that there is no one-size-fits-all solution in regard to the central clearing services being launched, but there is optionality in the market.
She stated: 鈥淪ponsored GC was launched a couple of years ago, and since the finalisation of the mandate at the end of 2023, it has increased more than 400 per cent, crossing over US$800 billion in daily volumes. This shows that the market is moving into cleared channels ahead of the mandate.鈥�
Jeff Sowell, head of Financing Solutions product strategy, North America at State Street, commented that the Sponsored model was launched by FICC in 2005 and, up until 2017, was 鈥渁 really niche鈥� model.
鈥淲hen that model was created in 2005, I don鈥檛 think anyone in their wildest dreams could have imagined that it would turn into the US$2.4 trillion wagon that it is today,鈥� he added.
He referred to the legacy Sponsored model of 2005 as the 鈥渏ack of all, master of none鈥�. A model which worked well through challenging market conditions, but one that does not offer the same efficiency as the new models which better fit specific client archetypes.
Collateral-in-Lieu is an extension of the Sponsored general collateral (GC) service, explained Pavlidou. The Fixed Income Clearing Corporation (FICC) takes a lien on the assets that sit on the cash investors鈥� triparty account, and because of this lien, the model is able to offer margin and capital benefits if clients are risk-weighted asset (RWA)-constrained.
This is possible because, in most circumstances, there is no need to collect margin on behalf of the Sponsored member, and there is no need to guarantee the performance of the trade, which can reduce RWA exposures.
Sowell added that this model is well-suited for institutional cash investors, who have a focus on yield, liquidity, and are more sensitive to counterparty credit risk. The model also provides the operational ease of having a triparty agent bank hold the collateral and that undertakes the allocation, sending, reporting etc.
The ACS is the next model BNY has launched and is an extension of the existing Agent Clearing Service (the bilateral version) that the FICC offers, except this new model is on triparty. Essentially, this model enables members to continue to receive the benefits of the service by leveraging the operational benefits of operating under a triparty structure, as well as the net margin benefits that the ACS model provides.
For Sowell, the ACS model is well suited for firms like alternative investment funds, which are focused on financing treasuries and minimising all-in cost as it is the most efficient buy side access model that FICC offers.
Looking forward, Sowell predicts: 鈥淲e鈥檙e going to see new balances come directly into the new models, and then existing activity will migrate out of Sponsored into those new models over time.鈥�
Moving onto a solution which has created much discussion, Pavlidou discussed Done-Away. She said: 鈥淔rom a Global Collateral standpoint, when we think about Done-Away, it is important to have the operational structure to support it, which we have built on our triparty infrastructure. As we get closer to the mandate, dealers might realise that they don鈥檛 have as much capacity to either clear or paper more counterparties, this is where Done-Away will be a solution that is very beneficial in helping the market comply with the mandate.鈥�
Readiness
The conversation shifted to discuss the current state of readiness of firms in both the US and Europe, with moderator Andrew Lazar, managing director, head of rates sales at Buckler 网爆吃瓜, asking: where do we find ourselves right now?
Alice Elizabeth Othigo, director, global product manager at BNP Paribas, noted: 鈥淭his regulation is a major shift for the industry and for the current market, and the industry is preparing for this change. Yet the preparatory steps and the readiness of different firms may vary depending on market segment, regional presence, clearing model options, and project headways.鈥�
In terms of the fundamentals that firms are working on in order to be ready for this regulation, Othigo notes a number of important factors. First, assessing trading patterns, identifying eligible transactions, engaging with trading counterparties and clearing houses to define the best-suited clearing model. Followed by legal repapering and streamlining the flow to avoid activity bottlenecks and drawbacks.
Then, running a comprehensive programme to adapt technology and operating models to the new environment and latest workflows features. As a result, delivering scalable and integrated service models supporting firms鈥� and their clients鈥� business strategies and goals.
For Andy Hill, managing director, co-head of Market Practice and Regulatory Policy at the International Capital Market Association (ICMA), it is as much about awareness as preparedness.
US Treasuries are traded globally, and for ICMA, the association has members in 70 different jurisdictions, all of which trade US Treasuries to some extent or another. It is no longer simply about whether these jurisdictions are getting ready, but whether they know they are in scope.
In terms of non-US members who are fully prepared, an ICMA survey from late 2025 reports that only four per cent are fully prepared, while around half have not begun or are just starting. As a result, the association is working to raise awareness.
The survey also highlighted a number of current challenges facing in-scope firms, the two key points remain around understanding the cross-border application and whether firms are in-scope. Other issues include coordination with counterparties, regulatory compliance and documentation, operational readiness, and choosing access models.
Notably, Hill highlighted a potential negative impact of the mandatory clearing mandate. He explained that from a non-US entity perspective, if US Treasury and Treasury repo is not part of a firm鈥檚 core business, then firms must ask 鈥� considering the operational lift, the documentational lift, and even the regulatory uncertainty of being in scope 鈥� if they need to use US Treasuries or if they can use something else.
As a result, he anticipates that a number of marginal users of the Treasury market will likely step out, due to the complication and expense of this regulatory move.
Further, Hill said there remains uncertainty around triparty in a European context. He explained that if a firm moves forward with a triparty which starts with US Treasuries as part of the basket, they are in scope. But if they start off without US Treasuries and these are later caught up in that triparty due to optimisation engines and dynamic collateral allocation, the regulatory requirement remains unclear.
In his conclusion, Hill stated that there is still much to work through from a European perspective, and in a global context in general.
It鈥檚 all about data
The Data Quality and Controls and Oversight in Regulatory Reporting panel explored how data is used in securities finance, how it impacts the decisions made by industry participants, and how it looks at risk management and compliance.
Gavin Marcus, head of North American sales at S&P Global Market Intelligence Cappitech, began the panel by explaining how the firm is evaluating data quality, highlighting that regulation is the key driver. He noted that feedback from clients, regulators, and industry bodies, suggests that firms are looking at data quality now as they cannot 鈥済et the house in order鈥�. He underscored the importance of analysing their data going back to source systems such as trade capture, how firms are booking the trades, risk systems, order management systems, and if the reference data is properly established.
Building on Marcus鈥� remarks, Thomas Veneziano, director of product management at Broadridge, said that data quality is now driving much of the industry鈥檚 decision making. He noted that with greater clarity around regulatory timelines, market participants have the opportunity to improve data quality, increase transparency, and engage more proactively with regulators on how the date should be used. Previously, tight timelines meant firms were focused primarily on meeting reporting requirements rather than ensuring quality data.
鈥淒ata is no longer a back-office or reporting issue or conversation,鈥� according to Tommy Ros, senior vice president, capital markets services at Delta Capita. 鈥淭here is operational friction that now moves into pricing, equity, counterparty confidence, regulatory standing, and reputational damage.鈥�
He noted that the lack of regulatory change has led to an emphasis across firms looking at the change in speed and scale of their current operating models due to data flow from front-to-middle-to-back office now being almost instant, causing traditional, manual controls to feel outdated, and raises the question of how can firms improve their data quality as the regulator is putting such emphasis on it.
Concluding this section of the panel, Marcus noted that he has had numerous conversations with clients that suggest that data lakes are 鈥渋mperative鈥�, with firms analysing their data across the front, middle, and back offices as well as for regulation. 鈥淔irms have put the data lakes in place so that all the data can be governed correctly, with all the data points and elements needed for processing, pricing, liquidity, and regulation.鈥�
Panel moderator Nancy Steiker, senior director, Global 网爆吃瓜 Finance Product Management at FIS, then posed the question: why does data quality matter?
鈥淨uality is extremely important,鈥� said Veneziano, noting this is true for a number of reasons. Firms rely on a wide range of analytics to assess counterparties and determine appropriate trading strategies. He added that, from a client onboarding perspective, several variables must be considered, including standing settlement instructions (SSIs), concentration limits, and proper investment guidelines.
鈥淗aving data in a proper, verified, clean, quality format is key, because if a firm experiences a fail today, it may face TMPG claims, CSDR penalties, reputational risk, and potential buy-ins,鈥� he said, explaining that data impacts reputational risk, market risk, and financial risk.
Ros emphasised the importance of data quality, saying that regulators are now requesting to know where data comes from, who owns it, and how it can be trusted. 鈥淭raceability is a huge point of emphasis today, which opens up a number of questions when looking at third party and AI-integrated models across third party regulatory platforms or in-house builds.鈥�
Shifting the conversation to AI, Marcus noted that 鈥渆very single client鈥� is asking how AI is being used in tools and how AI can be used to create efficiencies and cost savings. He said at S&P, AI is being embedded in all S&P solutions as well as being used internally in everyday processes, with everybody being encouraged to evaluate ways they can use agents or tools to embed AI in their processes while keeping a human 鈥榠n the loop鈥�.
He adds the caveat that guardrails are needed, that code cannot be installed on desktops, that it needs to be vetted by the IT and security teams. 鈥淭he last thing you want is to see client information on the public domain because someone has made a mistake,鈥� he explained, highlighting that AI implementation has to be done in a slow, controlled manner to ensure it is done correctly.
The panel then moved on to discuss governance, with Veneziano underscoring that governance is shared across the organisation, and that middle office operations play a key monitoring role. He added that governance requirements vary depending on the data type, but also noted that operational risk, trading and market risk, compliance, and internal auditors all play a role in reviewing data as well as regulators.
Marcus added that data scientists are now being employed as data stewards across the industry.
Commenting on best practices to ensure front, middle, and back offices work together, speaking from Delta Capita鈥檚 perspective, Ros highlighted that building an ecosystem across the post-trade world has resonated with their client base in the past, having the ability to integrate with various systems that clients are using, and streamlining it into one platform that can aggregate, normalise, and present the data in an easily-digestible format for users across the board.
Moving onto the risks associated with poor-quality data and fragmented data management Veneziano noted that both financial and market risks can arise. He explained that if data is inaccurate or poorly maintained, risks such as incorrect SSIs may occur. He stressed the importance of maintaining strong controls to ensure that settlement instructions remain accurate. Poor data, he added, can create issues across both new and existing transactions, leading to potential fines or penalties.
Ros added that, as the industry becomes faster and the flow becomes almost instantaneous, front-to-back office firms no longer have the luxury to fix downstream regarding errors they have made in trade transaction reporting, adding another layer of risk across the board.
The importance of data transparency was echoed by every panellist, particularly the ability to trace the lifecycle and identify where (if at all) an error occurred. From a reconciliation perspective, Veneziano said firms need to ensure that data is accurate, properly maintained, and sourced from reliable systems.
鈥淭racebility has moved from being a nice-to-have to an expectation in today鈥檚 world, especially in legacy and fragmented environments that many firms still have, but the regulators do not care about complexity. The feedback we have received from participants is that having a golden trail of data lineage and traceability is becoming a huge point of emphasis,鈥� Ros added.
Concluding the discussion, market participants heard that data is the new goal, with firms wanting to do less with more 鈥� data is the way for firms to differentiate themselves, create new processes, be more efficient, and increase revenues.
The agent lender perspective
Agent lending is only commoditised if participants run a generic, undifferentiated programme, with market leaders distinguishing themselves through customisation, collateral and capital optimisation, and smart use of technology and data to align closely with clients鈥� evolving financing and regulatory needs, according to panellists.
Titled: 鈥楤eyond the baseline: how agent lenders compete in a commoditised market鈥�, participants reviewed how firms leverage collateral optimisation, proactive regulatory guidance, balance sheet and risk-weighted asset (RWA) management expertise, and which service innovations resonate with beneficial owners and where the market is heading as competition intensifies.
Jon Whiting, vice president, head of international and fixed income agency trading at Fidelity Investments, laid the foundation for the panel through his perspective on differentiated models from a beneficial owner standpoint.
鈥淲hat we are trying to do is ensure that we are extrapolating as much value as possible from the parameters our clients provide, including risk parameters, programme parameters, and cultural parameters.鈥�
He emphasised the need to maintain stability to ensure the firm can make as many transactions as possible in addition to ensuring there is a level of connectivity with street-side firms so they know the challenges and can provide mutually beneficial solutions.
鈥淔rom Fidelity鈥檚 Agency Lending鈥檚 perspective, we are trying to ensure that we are building a sustainably scalable model. This means we are keeping an eye on costs and maintaining efficiency while being a driver of technology solutions for our client base and the wider market.鈥�
Leading the panel forward, moderator Brian Morrissey, executive director, Americas buy side trading services sales, securities services, at J.P. Morgan, asked the panel what collateral optimisation means to an agent lender, and from a beneficial owner perspective.
Mark MacNeill, head of US equity and corporate bond trading at eSecLending, answered, highlighting that as the market is growing and becoming commoditised, a lot of performance at the margin can be generated from managing collateral well.
MacNeill states that managing collateral well, particularly through accurate cash forecasting and efficient intraday decision-making, or even smaller improvements, such as timing cash investments and repos more effectively, can add 10鈥�15 basis points, which becomes significant when applied to large balances.
He also underscored that by understanding the needs of counterparties, firms can optimise collateral deployment and create more value on the lending side.
Alvin Oh, global head of securities lending product at Wematch.live, adds that, from a trading platform perspective, there is a need to introduce effective utilisation of inventory, and the question of 鈥榟ow do we allow agent lenders to earn more and lend more effectively on platforms such as Wematch鈥� needs to be answered. 鈥淚t is through analytics, workflow automation, and smarter matching tools that we can help firms maximise their lendable assets, reduce inventory, rather than lending up and having to finance it overnight, and bring about collateral that they can then run their models against,鈥� Oh concluded.
Moving on to answer the question of what clients are asking of agency lending programmes and discussing regulatory impacts, Whiting notes that clients are seeking preparedness for structural market changes such as T+1, Treasury clearing, and equities as collateral in the US 鈥� 鈥渃lients want to know what is going to impact their programme and their day-to-day as it currently stands鈥�.
Anthony Toscano, North American head of global securities lending solutions at MUFG Investor Services, shifts the discussion to the ETF market, highlighting that the firm serves a diverse client base, including large institutional investors and quickly growing ETF providers 鈥� many clients are seeking help on central clearing as simple as the buys and sells of their treasury positions.
Clients are also constantly wanting to revisit their bespoke programmes, and are looking to see what the implications of tweaking their current guidelines would be 鈥� whether broadening their collateral sets or changing buffers or the dynamic liquidity tests there are prior to putting a security on loan.
鈥淓TFs in particular, continue to grow rapidly, meaning the buffers that have been used may be slightly too conservative. Being a full service agent lender means, not only are we finding distribution channels for this supply, being adaptable to the collateral types to address the binding constraints of the counterparts, but also looking at the client guidelines and parameters to see if there鈥檚 things that can be tweaked,鈥� said Toscano.
From a J.P. Morgan perspective, Morrissey notes that clients are continuously asking that the agent lender sits more within the financing ecosystem, and they are looking for providers that are able to handle the full lifecycle of financing, such as liquidity management, collateral optimisation, and seamless connectivity to financing and margin workflows and revenue generation.
Furthering the discussion into service implementations, Oh comments on the use of a trading venue and electronic marketplace that allowed the firm to open up broader liquidity pools rather than relying on traditional bilateral relationships. He noted that this means clients can access a wider network of counterparties, therefore creating more opportunities for supply and demand and improving overall market efficiency in addition to providing market transparency.
Concluding the panel, MacNeill underlines the importance of agent lenders to consider the opportunity cost of a loan versus cash, or a loan versus non-cash. He also notes that for clients that primarily might be in overnight vehicles that can take equity collateral, this provides a good opportunity and may allow them to perform better from an intrinsic standpoint, and that is important to make sure clients see incremental performance.
Tokenisation moves from theory to infrastructure
The 鈥楤lockchain, Digital Assets, and the Tokenisation of 网爆吃瓜 Finance鈥� panel brought together perspectives from across the securities finance ecosystem, with participants largely aligned on one point: digital assets are no longer a theoretical discussion.
Moderated by Tina Joshi, head of securities finance and collateral management solutions at Broadridge, the panel explored how institutional thinking around tokenisation has evolved, what is still holding adoption back, and whether the industry is genuinely approaching a large-scale inflection point.
Opening the discussion, Joshi asked whether the industry is now at an inflection point or still firmly in the infrastructure build-out phase. Sasha Sitsker, vice president, solutions engineering at EquiLend, argued that the answer is both. 鈥淭here was a lot of apprehension about digital assets,鈥� he explained. 鈥淣o one knew exactly what stepping into the digital asset space represented, or what the benefits necessarily were.鈥�
According to Sitsker, the industry has now moved beyond that uncertainty, with early adopters and major market participants beginning to align around what future infrastructure may look like and the tangible benefits of its adoption. 鈥淎t least now we know which infrastructure we are building out to,鈥� he noted.
Nathaniel Lindsay, vice president and head of electronic trading, agency securities lending at State Street, agreed, but highlighted the importance of the evolving US regulatory landscape in accelerating institutional interest.
He pointed to the January 2025 executive order supporting digital assets in the US, alongside the US 网爆吃瓜 and Exchange Commission鈥檚 (SEC鈥檚) rescission of Staff Accounting Bulletin (SAB) 121 through SAB 122.
Lindsay explained that SAB 121 effectively prevented prudentially regulated institutions from engaging meaningfully in digital asset custody because it required assets under custody to sit on balance sheet. 鈥淚magine if State Street鈥檚 US$50 trillion was on balance sheet,鈥� he remarked. 鈥淚t is completely infeasible from a regulatory capital standpoint.鈥�
Steve Everett, chief commercial officer CDS and head of post-trade innovation at TMX Group, meanwhile argued that the infrastructure discussion needs to be viewed more holistically. While many digital asset discussions focus on innovation layers and settlement rails, Everett suggested the industry often overlooks the operational bottlenecks inside firms themselves. 鈥淣o matter how efficient your rails are, when a client back-end infrastructure does not quite meet up to the same speed, it all goes down to the lowest common denominator,鈥� he observed.
When asked about which part of the industry could see scale adoption of digital assets, Everett highlighted private markets as one area where digital transformation may accelerate more quickly because, unlike many public markets, there is little entrenched infrastructure to replace. 鈥淧rivate markets actually have very little infrastructure, certainly in Canada and it鈥檚 mostly paper based,鈥� he said.
A major theme throughout the panel was the growing distinction between cryptocurrency and broader digital asset infrastructure.
Sitsker noted that for many years digital assets were effectively treated as synonymous with cryptocurrencies such as Bitcoin and Ethereum. Now, however, the industry is developing a much more nuanced understanding of the space, and is better able to distinguish between the asset class and the underlying technology. This includes tokenised securities, tokenised real-world assets, stablecoins, and even the use of distributed ledger technology (DLT) purely for operational workflows.
As an example, Sitsker referenced EquiLend鈥檚 1Source distributed ledger platform, which creates a single onchain version of a securities lending contract accessible to both counterparties. 鈥淚f adopted at scale, it completely eliminates the notion of counterparties being out of sync or needing to come and reconcile their contracts on a day-to-day basis,鈥� he said.
Lindsay added that State Street鈥檚 strategic approach to digital assets has also evolved significantly. 鈥淥ur 2026 gross technology investment in digital has grown orders of magnitude since 2025 and even more substantially since 2024,鈥� he revealed.
According to Lindsay, digital assets have shifted from being a niche custody discussion to something being assessed strategically across the entire organisation.
The question of complexity versus simplification was another central topic during the discussion.
Asked whether tokenisation simplifies custody operations or makes them more complex, Lindsay acknowledged that the short-term reality is likely more difficult operationally.
Maintaining synchronisation between traditional securities and their digital representations, he explained, introduces challenges around minting, burning, corporate actions, stock splits, and dividend processing. However, he argued that in a fully digitally native environment, complexity could ultimately decrease significantly. 鈥淭hat is where we start to reap some of the more compelling benefits of this technology, like programmability and atomic settlement,鈥� he noted.
On the question of what still prevents institutions from moving faster into digital assets, the panel broadly agreed that the primary challenge has shifted away from regulation and towards technology integration.
Sitsker argued that the industry now broadly understands the regulatory framework, and the focus has instead turned towards scaling adoption and reducing integration pain points. 鈥淩ight now everyone has got their sights focused on: okay, we know what the rules of the playground are,鈥� he said.
Lindsay similarly highlighted the importance of education and institutional-grade security. He explained that many clients still associate digital assets primarily with cryptocurrencies, rather than with tokenised representations of traditional financial instruments. 鈥淲e think about abstracting away complexity and really being a bridge between traditional and digital markets for our clients,鈥� Lindsay said.
The concept of institutional-grade infrastructure became a recurring theme throughout the discussion. From a vendor perspective, Sitsker argued that many of the industry鈥檚 early assumptions around decentralisation are now being reconsidered.
Rather than prioritising decentralisation, he suggested institutional markets instead require scalability, security, and trusted operators. 鈥淭he decentralisation angle was probably more of a hindrance than a benefit to an industry that needs clarity around who is maintaining these networks and applications,鈥� he explained.
Everett agreed, referencing Canada鈥檚 Project Samara as an example where assumptions around disintermediation did not ultimately hold up under scrutiny. 鈥淭he starting point should not be disintermediation,鈥� he argued. 鈥淭he starting point should be: what value is being derived and what exactly is the problem we are trying to solve for?鈥�
The panel also challenged some of the industry鈥檚 assumptions around atomic settlement.
While often presented as one of tokenisation鈥檚 key advantages, Everett questioned whether marginal settlement speed improvements alone justify the operational transformation required.
鈥淚s atomic settlement as big a value as we say it is?鈥� he asked. 鈥淭oday settlement in traditional infrastructure is measured in 10ths of second. The issue isn鈥檛 the technology, it鈥檚 more about how the ecosystem connects with each other today and the associated behaviour which is concomitant.鈥�
Instead, Everett highlighted programmability, reconciliation reduction, and cross-border collateral mobility as potentially more meaningful long-term benefits as opposed to pure atomic settlement as a value driver. 鈥淐ross-border mobilisation of securities is really not great,鈥� he said, suggesting digital infrastructure could materially improve that process.
The discussion later turned towards competition and market structure.
While the panellists generally agreed that trusted incumbents are likely to remain dominant in institutional markets, they also acknowledged that barriers to entry are falling rapidly.
Everett suggested advances in technology and artificial intelligence are making it significantly easier for new firms to build digital infrastructure. 鈥淭he barrier to entry dropping means that you are going to naturally get more competition,鈥� he said.
Still, both Lindsay and Sitsker argued that network effects and institutional trust will continue to favour large incumbent providers, particularly in custody and asset servicing.
Finally, Joshi asked the panel which asset classes they believe will first achieve genuine tokenised scale beyond pilot programmes.
Sitsker pointed towards stablecoins, describing them as the most easily understood digital asset use case and potentially foundational infrastructure for digitally native transactions.
Lindsay instead highlighted tokenised money market funds, citing growing institutional demand for yield-bearing digital cash instruments that can also be mobilised as collateral. 鈥淭he product-market fit is pretty clear,鈥� he said.
Everett agreed with the money market fund thesis as collateral, but also argued that private markets may ultimately see the most complete digital migration due to their lack of entrenched infrastructure. 鈥淲hen that (private markets) moves across to this technology,鈥� he concluded, 鈥淚 think it is going to move at scale.鈥�
Away from the silo
The 鈥楥ollateral Management and Optimisation: Making the Most of What You Have鈥� panel highlighted one clear message: collateral optimisation is no longer just a back office efficiency exercise.
Moderated by Amy Caruso, head of collateral initiatives at ISDA, the discussion explored how firms are trying to put the right collateral in the right place at the right time, while navigating constraints around capital, return on equity, liquidity, and operational complexity.
Opening the discussion, Caruso noted that collateral optimisation can mean different things to different firms. Some focus on capital, others on return on investment, and others on ease of movement.
For Joseph Pirro, director and head of equity lending and repo Americas at ING, the answer has been a major internal overhaul.
Pirro explained that ING鈥檚 securities finance division had historically been split across fixed income repo, equity lending and equity repo, and linear equity derivatives, with each area managing its own funding and collateral.
That siloed model has since shifted towards a more centralised approach through a Funding and Liability structure at ING, which has now been in place for many years with Global 网爆吃瓜 Finance. The project is a bank-wide initiative to make Global 网爆吃瓜 Finance the hub for all collateral within the bank.
鈥淲e are bringing it all back and doing it all from an optimisation perspective, but it all starts with the data and making sure that it is clean,鈥� Pirro said.
He added that the project is still in its early stages and is expected to take three to five years, with a focus on dashboards, system connectivity, front-end booking systems, and reducing manual intervention.
Ted Leveroni, head of margin services at BNY, said buy side firms face a similar problem, even if the details differ. 鈥淭he problem statement, I think, with buy side and sell side, is universal and similar,鈥� he said. 鈥淚t is inefficient operations leading to inefficient use of assets, leading to a drag on performance.鈥�
For Leveroni, the first step is an aggregated view of assets, followed by optimisation tools and mobilisation capabilities. He said firms need to see not only their long assets, but also collateral received through financing and derivatives activity.
Ed Corral, head of collateral services at Pirum, argued that vendors can help firms accelerate this process by reducing internal friction and supporting consolidation of inventory and eligibility schedules.
Reflecting on his own experience, Corral said optimisation can take years, but the value can be substantial. 鈥淲hen you finally get there, the value is incredible,鈥� he said.
From a central counterparty perspective, Harsh Devpura, director, securities finance at OCC, highlighted the role of infrastructure modernisation.
Devpura pointed to OCC鈥檚 Ovation programme, describing it as an investment to comprehensively redevelop and modernise the company鈥檚 risk management, clearing, and data systems.
According to Devpura, the platform is moving from a single mainframe to microservices in the cloud that can support faster product delivery, optimise workflows, provide real-time data that can act as tools for members to develop their own applications, and enhanced security and improved resiliency.
He also highlighted OCC鈥檚 exploration of new membership access models and tri-party services.
The panel then turned to artificial intelligence, with speakers broadly agreeing that the technology could have a significant role in collateral optimisation.
Corral continued that optimisation is particularly well suited to AI because many of its objectives can be represented mathematically. 鈥淓very complex objective in optimisation can be represented as a formula, so it is tailor-made for AI,鈥� he said.
Pirro said ING is already using AI for admin and to help with on the desk coding projects. He identifies possible future use cases the market could consider, such as digitising agreements, reviewing collateral support annexes, and summarising regulation.
Leveroni suggested that AI could ultimately support pre-trade optimisation, helping clients identify the best source of financing for a particular security at a particular time. However, he cautioned that firms still need the underlying data, connectivity, and asset mobilisation capabilities before they can fully benefit.
Devpura added that firms can use analytics to assess what their margin requirements may look like before portfolio finalisation.
The use of non-cash collateral was another major theme.
Leveroni said the driver behind greater use of securities as collateral is simple: 鈥淚t is cash drag.鈥� He argued that most collateral support annexes remain cash-only, creating a need for standardisation, automation, and more efficient ways to update documentation at scale.
Pirro agreed, noting that equities and Treasuries have long been used in securities lending, but that large pools of buy side assets remain idle. 鈥淭he idea of being able to mobilise a lot of these things that were idle, even if you just mobilise it another step, is a step in the right direction,鈥� he said.
Corral recalled previous efforts to post non-cash collateral under existing agreements, only to find that counterparties were not operationally ready to accept it. 鈥淭here was nothing more to it than that,鈥� he said, recalling one operations manager鈥檚 explanation that cash was simply easier to post.
Finally, the panel considered tokenisation and its potential impact on collateral management.
Leveroni argued that tokenised collateral could support intraday margin calls and ultimately better pricing. 鈥淭o me, you are taking an inefficiency out of the market, which translates into cost out of the market, which is better for the investor,鈥� he said.
Corral highlighted the value of models that link traditional collateral with tokenised representations, suggesting these hybrid structures may be more realistic in the near term than a fully digital replacement of existing markets.
Devpura said tokenisation could also support extended trading hours, particularly in a 23/5 or 24/7 market environment. 鈥淚f you really want to get into a full 24/7 or 23/5, tokenisation becomes a strong workflow tool,鈥� he said.
Across the discussion, the message was consistent: collateral optimisation requires technology, but it begins with visibility, clean data, and the ability to mobilise assets across fragmented systems.
NO FEE, NO RISK
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to 网爆吃瓜 Finance Times
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to 网爆吃瓜 Finance Times
