Research Reports » Clearstream June 2019

Roundtable: The transparency trap

Our panel of experts discusses the best ways to achieve efficiency, data regulations, the challenge of fragmented market practices and the future role of technology.

Transparency_roundtable

Bernard Tancre (head of Investment Fund Services, Clearstream)
Gautier Despret (client director, IQ-EQ)
Keith Hale (fintech investor, non-executive director & adviser)
Mark Gem (chair of Clearstream Risk Committee, Clearstream)
Rafal Kwasny (vice president, international transfer agent EMEA, Franklin Templeton)
Revel Wood (independent director and consultant)

Funds Europe – How can the demand for transparency be balanced with the need for efficiency in fund distribution?

Bernard Tancré, Clearstream – As a central securities depositary (CSD), Clearstream is focused on settlement, efficiency and safety. When we look at investment funds, we question why we treat them in a different way to other financial assets. The big difference is the amount of information that flows through the parties of the transaction. For example, the issuer needs information about how its salesforce is performing. There is also the regulatory aspect via AML [anti-money laundering] and KYC [know-your-customer] and the demand from fund providers to have as much transparency as possible and to then check every name from an AML/KYC standpoint. I understand the reasoning but would challenge it. We want to separate the information that needs to be processed from the information the asset manager wants to use for commercial or compliance reasons.

Revel Wood, independent director – There are different needs and demand from different parties. There is also a disconnect between different regulators and their treatment of this data. The FATF [Financial Action Task Force] states there must be a complete look-through to the ultimate asset owner while Esma [the European Securities and Markets Authority] has stopped short of that demand. And then you have fund promoters with their trailer fees that want to know what the distribution chain is costing them and the sales and marketing teams that want to know who’s buying funds in which markets. The other side of the transparency debate is the efficiency and I struggle to see how we can achieve full efficiency without cross-industry platform initiatives. In the UK there are banks acting as placement agents, IFAs [independent financial advisers] and third-party marketers, all with different forms and with different local regulations. Europe is not like the US, where all 50 states have the same basic process and rules. This is where the likes of Clearstream could help.

Keith Hale, fintech investor – Does more transparency all the way through the value chain create more efficiency?

Wood – No, I don’t think it does, not until standard formats are established as was done in middle-office and settlement processes.

Rafal Kwasny, Franklin Templeton – I think transparency may actually lead to more efficiency. For example, everybody is collecting KYC documents and performing AML processes in the same way and the whole industry sees it as a burden. Sooner or later there will be a solution there that will lead to reduction of paper flows and a more efficient model. I do not see that happening very soon but many people are already working to find a solution. Another example is the transparency around costs. Since 2019, there will be more transparency around cost as a result of MiFID II and that may lead to further cost pressure and greater efficiency.

Hale – If you were starting the investment value chain from scratch today, you would have far fewer intermediaries with more inherent transparency, thereby reducing the need for reconciliations between the various parties such as omnibus accounts and individual investors. So I don’t think transparency and efficiency are balanced today, given the current structure of the wider industry.

Tancré – We are looking for this new infrastructure to help with AML/KYC, but a lot of asset managers are already using distributors and paying them a lot of money. One of the things they can do is those AML checks. When we are an element in the chain, there’s the desire to look behind rather than really trying to trust the nominee who is doing the work. It’s ironic to me that we’re trying to find new parties to whom we can delegate but such delegation is already inherent in the correspondent banking model. That is why this desire for transparency is really motivated by commercial and not regulatory reasons. The problem with the latter is that it very often gets in the way of efficiency and generates additional cost, because all this information needs to flow through the channels that are normally just for the transaction.

Gautier Despret, IQ-EQ – We do not all speak the same language when it comes to transparency. What is requested by the fund manager is different to what is requested by the regulator. They do not want or they do not ask exactly the same thing, so we are sacrificing efficiency for the sake of transparency. We need to find a way to reduce the amount of paper that is involved.

Mark Gem, Clearstream – In all other financial instruments, if you have a value chain where the intermediary is itself equivalently AML-regulated, then you just need to deploy risk-based controls. These controls are not aimed at transparency in the sense of information or documentation, they’re aimed at making sure that you as an upper-tier intermediary know if there’s a problem downstream to which you should react by, for example, not processing or whatever it may be. Nowhere else, from bonds to equities, does this notion that the investor’s identity needs to be known at all steps in the chain ever come up.

Funds Europe – Should fund companies be careful about how much customer-level data they receive because of the regulatory responsibilities they might incur?

Kwasny – Most of the asset managers do not want to own the end customer. They want to know end-investor product needs and their buying behaviour but not necessarily collect their personal identification details. MiFID II [the second Markets in Financial Instruments Directive] rules around collection of data related to product governance do not require sharing all the details such as the beneficial owner’s name, address and so on.

Tancré – That’s why MiFID II is really welcome from our standpoint, because it provides a good regulation justification and, if you like, leverage to have the commercial information, and the right level of commercial information that you want to have anyway, so that you don’t have to call to the AML/KYC which is generating this friction in terms of efficiency.

Wood – This is incredibly complex because there’s a large remit of regulations that are sometimes conflicting or different in nature, from the SEC [Securities and Exchange Commission] to MiFID II to AML and KYC to GDPR [General Data Protection Regulation]. The divergent nature and purpose of some of these regulations make it difficult to know how much information and for how long you keep information. It is a problem that we have not solved. We have said over and over again that the fund company, and fund promoters especially, do not need all that client data. As the management company – and here it’s been difficult over the years, because the market hasn’t been harmonised – we have seen challenges where regulation applying to a management company may contradict target market regulators’ data-protection laws. But we still need to ensure the distribution chain is being overseen and that there is equivalent regulation for our delegates and global distributors.

Hale – It depends on what you mean by ‘customer’ – the end investor or a private or retail bank that is servicing the end investor? Or the fund platform in between? Or any of the other intermediaries in the value chain? From a commercial perspective, the fund wants as many inflows as possible, as well as catering for the necessary regulatory requirements. But the actual end customer is often not clear in what is a multi-layered hierarchy. Currently there is very little data used by the fund manager compared to what is available at the end-investor level. The fund managers don’t need complete transparency but I understand why they’d want more understanding of the end clients than they have today, and not just for regulatory purposes.

Tancré – There’s an optimum level of data, depending on the distribution model, that will satisfy my commercial needs and no more than that.

Kwasny – It’s not a question of whether we should avoid collecting data because we are afraid of not complying with GDPR. It is instead about whether the data can actually bring extra value and deliver better products and solutions and better client experiences.

Tancré – It’s a question of what you get exposed to and the ambition of what you do with it. OK, yes, you have that transparency because you feel you need it, but then you’re not consequently doing all the things you need to do with it. That’s where the risk is.

Funds Europe – Market practice varies significantly between markets. What are the main challenges involved with this fragmentation?

Despret – One of our main challenges is regulatory surveillance. There are differences between the Luxembourg, French and UK markets, even if the spirit of the regulations is broadly the same. Once rules are translated into different country law, some of that spirit is lost or diluted.

Wood – It will be interesting to see how the market evolves. One of the consequences of the Retail Distribution Review in the UK was the consolidation of IFAs. We are already seeing consolidation in some of the management company space as a result of increased substance requirements. Surely at some point, there must be a utility because we all individually collect much of the same information, and I don’t believe this creates competitive advantage but does create cost and inefficiency for the end investor. But then the issue of GDPR comes up and how personal data is kept by the central hub but used by multiple parties. I’m sure some smart disruptor firms will figure it out because we have to have more efficiency, transparency and value for money for end investors in an increasing regulatory environment.

Hale – That’s the irony of the whole AML situation. Generally speaking, too much time and effort goes into AML and checking transactions that are not suspicious. In fund distribution, you’ve got the same four or five thousand distributors that everybody uses and the industry puts untold duplicative effort into checking those accounts and transactions by all the TAs and asset managers. Why don’t we just have a centralised capability for low-risk transactions and then focus on those peripheral cases where there are dubious-looking entities and transactions?

Gem – Yes, I know and that’s exactly the point I’m making, it’s diverting that effort.

We’d strongly favour some kind of repository of information that embraced all of the intermediaries and participants in the system, because clearly that’s what has to happen for that mission to be fulfilled. It’s not rocket science to collect the documentation. One thing that will convince the regulators is if the documentation is provided by the firm itself, not by a data vendor pulling it off the internet. We could do that too. That, to me, would take a lot of pain out of this, and I think even the most risk-based compliance teams are probably only spending 25% of their resources on risk-led duties and the rest on rules-based measures. Compliance teams have a strong incentive to avoid audit and regulatory findings. If you’re not collecting the right bits of paper, the auditors may find that you have “structural AML deficiencies”.

Kwasny – The market fragmentation leads to lots of customisation. There are many market-specific operating models which require you to employ local service providers. In theory, you may be able to do it centrally but in practice, it doesn’t make any sense because of cultural reasons, for example. There are plenty of differences in distribution operating models. Some markets are more IFA-focused, some are more dominated by banks or nominee accounts, some are CSD [central securities depositary] markets where you see a different model. That creates extra cost, extra effort and a lot of customisation.

Gem – If you were building it from scratch, you wouldn’t do it in this fragmented way. But we are where we are, so let’s try and find ways to harmonise and develop a standard, because the current market is inefficient, costly and doesn’t really benefit anybody, certainly not the end investor.

Funds Europe – What role should blockchain technology play in fund distribution?

Hale – I’m an investor in a blockchain company called Cygnetise, because I am actually a little bit of a sceptic. Blockchain is a great technology, but it reminds me of the World Wide Web in the late 1990s which resulted in the dotcom boom and bust. There is a blockchain gold rush going on right now, but the real challenge is adoption. Many blockchain initiatives require replacing the current intermediaries to create peer-to-peer communication between parties. However, those intermediaries, including various banking entities, have very high walls and very deep moats due to the nature of their business. In other words, they’ve got customers, capital, incumbency and competency and lots of different reasons, good and bad, as to why they are there. As a result, they won’t be easily replaced without a high level of cost and risk.

Gem – And they’ve invested in the blockchain companies.

Hale – A lot of blockchain propositions like FundsDLT, FNZ, Calastone and Iznes are attempting to completely change the way that fund distribution works today. These propositions are often based on a single blockchain-based registry that everyone shares. I think it’s a commendable concept but the adoption challenges are enormous, due to replacing the complexity and disintermediated processing in the current model servicing trillions of assets in Europe.

Kwasny – I would agree. Blockchain has the potential to impact the distribution of funds only if distributors start using blockchain, but I don’t hear much about distributors implementing it. There is this concept of pan-European transfer agent, but it’s not about distribution, it’s about a new system developed in the new technology rather than a change of the fund distribution model.

Gem – What is transforming parts of financial services distribution? It’s certainly not blockchain. Twelve months ago, if you put blockchain in the name of your equity, you got a 40% increase in your stock price in the next seven days. Today, that figure’s down to about 17%. We monitor this. Where you can see the transformation actually taking root is in the notion of open banking under PSD2 [the revised Payment Services Directive], where you force bank account providers to make their information available to third parties – fellow banks, but also certain specialists or disruptors. That is indeed transformative in areas like retail FX, in person-to-person payments and so forth. The technologies that enable that, interestingly, do not include blockchain. The major enabler is the application program interface (API), because it is technology that allows you to make that data available to others.

Hale – The funny thing about APIs is that they’ve been around ever since I was a programmer in the 80s. It’s just that initiatives like Payment Services Directive II have made them fashionable and more relevant to people other than technology geeks.

Kwasny – Blockchain is a technology. It doesn’t revolutionise business models on its own. It requires scale and volume, and a lot of cost must be added to build the functionality to make it multi-asset and multi-markets-compliant. The question is whether it will sort any of my business problems. If I replace one system with the other, it’s just doing the same in a different technology than the current system. Functionality-wise and from an end-customer perspective, it doesn’t sort any of the issues. Distributors still want to send Swift instructions and receive Swift confirmations back.

Hale – These initiatives only work if you manage to get an industry-wide utility off the ground, and to get utilities established is notoriously difficult, because it means removing the high walls and deep moats of the incumbent intermediaries. I’m not sure there is a real business case at this stage.

Tancré – There is a business advantage indeed when you look at the end state. It’s the road to that end state that has a difficult business case.

Gem – What adds value is first standardisation, and second the utilities and technology that enable it. Standardisation is only very rarely the limiter. I can think of specific cross-industry processes where even if you told me that I was not allowed to use any technology invented in this millennium, I could take 95% of the cost out, but only if all of the participants cooperate. That’s where the value is, but the problem is of course that your cost is someone else’s revenue and so forth. A lot of people, including frankly all of us, except maybe Keith, live from friction.

Despret – I think blockchain could create some kind of revolution, particularly in the alternatives world. We need to separate the fantasy from the reality, because we know that we will never have the full disintermediation. We will still need someone to validate the transaction, especially if the transaction is with real money and not with digital assets... But with some asset providers, we think that blockchain could bring something to the industry, maybe not for the fund distribution but more for the transfer agent by adding smart contracts, by just having software in place for the AML/KYC perspective, by just replacing low-value tasks and helping people just to focus on the high-value elements. That’s why blockchain could also be helpful, but the issue with blockchain is also the fact that the data is public. This will create some data security issues or restrictions on who will have access to what. Potentially you could have a fund-specific blockchain or a transfer agent blockchain within the blockchain.

Hale – There’s no issue with having a private blockchain. The challenge is the industry-wide adoption. To create real value, it needs to become a utility that the industry uses. I get the fact that if you’ve got a distributed ledger where all players up and down the value chain or across organisations share in the same registry, then there is huge benefit, rather than everybody having parts and copies of the data then reconciliation between them. I think we may see small pockets of blockchain activity appearing in specific regional markets or in an asset class like private equity. But to get it fully adopted in the mainstream is highly challenging, so will take many years.

Kwasny – Various initiatives and industry projects try to do the very complex things in blockchain rather than simple peer-to-peer transactions or sharing public information like market data or of funds-related static data. It is currently a huge spaghetti-like model, so why not put it on the blockchain? Everybody is complying with the same rules, so there is a potential to centralise it. Whether it’s data providers or a fund or an insurance business, whoever needs the data, if we would all take the data from a single source, then it could be a single source of truth. To me that’s a perfect example where the blockchain business case could be built.

Gem – Why are we so much worse at blockchain than our fathers were? What I mean by that is there are several instances of the industry, particularly in the 70s, where competitors got together and built single points of truth in order to solve real-world problems. Our firm [Clearstream] is a product of that and Euroclear is another. Virtually every single domestic capital market organised a depository function only in the 70s. In other words, when the operational pain of the spaghetti got to a certain point, our fathers grasped for the blockchain, which in those days was obviously more technologically primitive but amounted to the same thing – a single record. Why was it easier then than it is now? Is it just simply that the pain is not enough?

Hale – No, it’s the cost and risk of changing spaghetti that has been added on to their legacy technology developed in the 70s for storing a single record to deal with the ever-more complex environment we have developed since then. For example, back in the day, asset managers used to have a registry of retail investors. Today we have a highly complex multiple-layered value chain.

Gem – Yes, the settlement agents walked around the City of London with briefcases.

Hale – Now we have a depositary doing the settlement and we have a fund platform with the global distributor talking to a local distributor. The world has got more complicated. There are bigger assets and more complicated, international flows. Because of that legacy, to then reinvent it in a standardised, simplified blockchain-based solution is going to take some time.

Kwasny – There is also people’s resistance to change. Even the simple changes have not been fully implemented, for example, moving to Swift ISO20022. There are still people using ISO15022. And there are still companies sending faxes.

Gem – I asked the same question to Ian Saville who set up Crest in the UK. He said it wasn’t just that it was painful and it wasn’t just that it was expensive. What triggered people into action cooperatively was the fact that they could not physically get the work done any more. It was the same with the foundation of the DTCC [Depository Trust & Clearing Corporation], because the settlement backlogs were running into months. Maybe you need market failure before you can implement big market changes.

Funds Europe – What technologies, other than blockchain, will have a transformative effect on the fund distribution market in the next 12 months?

Hale – The lowest-hanging fruit is robotic process automation (RPA). There is a huge opportunity to use robotics to solve some mundane, manually processed problems and increase efficiency. You can make some simple fixes to manual processes using RPA technology.

A more interesting subject is the development of robo-advice. At the moment, we are in version 1.0 in terms of the artificial intelligence and machine learning behind robo-advice. But if you can add a more sophisticated layer of machine learning that enables robo-advisers to augment what asset and wealth managers do in terms of generating alpha for their investors, that would be very beneficial. The technology already exists, so it does not involve reinventing the world industry but just doing a better job with better tools.

Kwasny – There is a transformational opportunity with the cognitive tools and artificial intelligence. In financial services, we are quite behind what is the norm in other businesses.

Gem – Robotics has been quietly transformational by keeping costs in check in a way that you can’t see. We’ve had amazing experience with robotics. I’m always been surprised by how primitive the actual technology of AI machine learning still is.

Basically, we use the same algorithms as Tesla uses in its self-driving cars to work out whether there’s problematic material in a prospectus.

Quantum computing is going to be the next thing, because it will make AI and machine learning actually meaningful beyond just flexible algorithms, which is essentially what they’re doing now. I think we’re going to start seeing the first edges of that into financial services this year.

Tancré – The danger with robotics is that it hides the lack of technology standardisation. My hope in this short 12 months is that we see more collaboration and adoption of existing standards by a wider group. This would significantly increase efficiency. It’s a pity that sometimes we have to use artificial intelligence or robots just to offset the lack of adoption of ISO20022.

Despret – Artificial intelligence will not transform the world tomorrow. Everything we teach the technology will be dictated by the market and its rate of adoption. It will depend whether other parties or customers are ready to have that technology or are tech-averse. There will be no revolution in the short term, but it will be a different world in the longer term. We took decades to digest the current regulations, so I would assume that we will take decades to move to a new world.

©2019 funds europe