Originally published 17-September-2019 on Linkedin by Gareth Nicholson, Head of Fixed Income DPM at Bank of Singapore.
In a effort to stay on top of Changes in the Bond market, below is a collection of ideas affecting the way the industry is changing. Some ideas are new, some are being revisited but pretty interesting time to be involved in markets revolution.
I would love to share ideas, so if anyof this interets you do reach out for a chat or a coffee.
3 BIG FINANCIAL INDUSTRY DISRUPTORS ALSO REINVENTING BOND MARKET
Artificial IntelligenceDistributed Ledger/ Block chainOpen Banking
Artificial Intelligence
There’s virtually not a field that Artificial Intelligence hasn’t touched, and the financial industry is no exception. Machine learning algorithms are particularly adept at analyzing huge amounts of data and finding patterns that would go unnoticed to humans.
Already, hedge funds and stock exchanges are making use of AI algorithms to become more efficient in forecasting the stock market and make smarter decisions. Numerai and Sentient are two Silicon Valley–based hedge funds that are challenging the giants of Wall Street with their AI-powered trade systems.
AI is also finding its way into banking. An Accenture survey of 600 top bankers found that AI will become the main way banks interact with their customers in the next three years, easing the workload for employees and enhancing the consumer experience.
Artificial Intelligence can also play a pivotal role in preventing fraud in banking and other online payment systems. E-commerce fraud prevention company Riskified is using deep learning and behavioral analysis models to enhance the process of stopping fraudulent transactions. This means the company is better positioned to find intricate fraud schemes while reducing false declines, a phenomenon that is no less damaging than fraud itself.
But for Bond Markets, where we are focusing here there are many exciting developments. Over the past five years, we have witnessed profound changes in the fixed income marketplace with counterparties increasingly adopting quantitative investing and liquidity risk monitoring techniques. These include systematic alpha and algorithmic trading, liquidity risk management strategy and reported thresholds, merging of fundamental discretionary and quantitative investment styles, consumption of increasing amounts of alternative data, and adoption of new methods of analysis.
AI Application for Bond Markets falls into 3 key areas:
Intelligent automation, which focuses on the intake of both traditional fundamental data and alternative data (.i.e. past issuance pricing across peer groups, timing vs. size, price prediction, price tension based on market sentiment ) and use this information in unison with portfolio management tools to better isolate risk-return trades.
Another key benefit is the ability to scale coverage and increase analysis speed using Machine Learning (ML) to test correlations on large issuer coverage universe, reducing the required resources and time (cost) and improve precision (revenue). Don’t forget the Bond universe is multiple times bigger than the equities and always changing given bond mature and get re- issued. Also given the Asymmetric nature of the instrument, with a relatively capped upside and zero downside, diversification is key which makes the ability to scale coverage in a targeted way very attractive.
The Over-the -counter nature of the fixed income market also offers large opportunities for disruption. The fixed income market is highly opaque, driven by disparate, manual communication channels that carry high costs for all stakeholders. Over the last few years we have seen big changes with the introduction of more digitized dealers, thus reducing costs significantly and increasing network connectivity by implementing a more versatile infrastructure through a combination of these Fintech providers and traditional channels.
Whilst I’m confident AI will continue to improve the execution and liquidity management part of Fixed Income investing, this topic has been well covered and I would rather focus on two additional benefits gained from more market data.
Enhanced decision Making and Smart Alerts is something we are exploring actively. Where 1,000s of bonds are monitored in real time checking liquidity, pricing, correlations, and volatility are key events and prompting Portfolio Managers of potential opportunities which may have been missed. Combined with proprietary data from in-house trade flow, AI can infuse the data into models to better understand client preferences and buying patterns.
Advanced Real-time pre and post risk management solutions offer an additional level of security expected from bond markets. Don’t forget that Bond markets are loved typically by an aging demographic given its predictability and lesser surprises. Risk management tools which reduce uncertainty even further are always welcome. For instance, in taking of alternative datasets with machine-learning algorithms can improve the coverage and robustness of risk models, as well as improve the quality of the data intake.
Before sharing some exciting AI solution providers in the Bond Space, we should touch on arguably the biggest challenge - Need for centralization of information
There is a great need for a fixed income big-data centralization where advanced analytics such as price discovery, liquidity risk management, intelligence gathering, pre-trade and post-trade analytics can be performed – to increase the overall efficiency of the fixed income market and understanding of the credit risk valuations. With no centralized hub, issuers and investors operate with partial awareness. AI application utilizing deep historical data records of fundamental data elements (audited statements, dealer supplied primary bond price quotations etc.) and secondary market bond trade points can solve this problem, but a centralized big-data hub powered with AI capabilities is key.
The buy side is spending more time on data strategy. They want to apply machine learning, intelligence augmentation, and at some point, A.I. But first they need to acquire and store the data. Buy-side portfolio managers want to know what bonds are liquid, but this is difficult information to come by given the OTC nature of the market.
Vendors such as Market Access and Bloomberg ALLQ offer solid platforms that aggregate liquidity and provide indications to buy sides. But for now these remain mainly focused U.S. markets, with limited debt on Emerging Markets.
One provider which I really like is Algomi, given it lets BUY-Side managers take back the ownership of the terabytes of bond data received through runs, axes, messages, emails, IBs, etc. Algomi ALFA provides cross-market information on liquidity and trade intent in order to give the buy-side a real-time view of the entire bond market, including government bonds, investment grade, high yield, emerging market, municipal debt and structured credit. This can increase the likelihood of a successful trade and this has the potential to be a ‘game changer’ for price discovery in bond markets. More informed decision-making through capturing a range of “Liquidity Signals”.
Below is a few more AI in Bond Market used cases
Recommending Bonds for Investment/AI Advisor
Synechron’s Cognitive ML Accelerator focuses on Product Recommendations for Fixed Income by drawing on historical bond trading data to uncover patterns and then recommend bonds that have similar attributes preferred by a particular trader. The engine pulls in, catalogues and analyzes historical Trade Reporting Compliance Engine (TRACE) data and can expand to include other data sets like MiFID II post-trade best execution data to enhance global, real-time portfolio management and trade decisions.
BondValue is a Bond centric platform looking to solve much inefficiency in the bond market by leveraging innovative technologies. I was fortunate enough to be a judge in the Fintech Shark tank 2018 where they blew the competition away. This shop is embracing all, including: 1) AI-powered News Engine that categorize bond news to reduce noise & gain insights using Natural Language Processing. 2) Bond Prospectus Scraping, including Covenant checks to compare bonds in minutes and make better decisions 3) Blockchain-based Bond Trading Technology
Remember it's not Data Overload, but Filter Failure.
Smart Desk Assistants.
The same kind of AI that Netflix uses to recommend movies to its customers are now being used in Investment banks to help target sales. Using AI analytics in a bond trading setting will eliminate much of the guesswork that’s currently used. AS markets open more, traders can see exactly what customers bought the previous week, which customers trade with which banks, those they trade with most often, and what securities they trade. Dutch bank ING has been trying out this approach via an AI tool called Katana. JPMorgan Chase has also been playing with AI tools in order to help traders and sales staff predict how markets will flow. This is not a new fad, but the soon to be BAU
Alliance Bernstein virtual assistant Abbie can suggest the best bonds to buy and sell based on pricing, ease-of-trading and risk. Unlike any human, she can scan millions of data points to filter the universe of outstanding bonds in seconds and identify potential trades to portfolio managers using other electronic tools the firm has built. Abbie 2.0 will identify bonds that people may have missed and will be able to spot human error and communicate with chat bots like herself at other firms.
Putting theory into practice
Bond desks have always attracted Math’s geeks, number guys that love details. Not surprising to find many mathematicians, engineers, physicists as portfolio managers, but only a small number get the space to explore some of the theory in practice, especially in traditional banks where quants fall more into risk than searching for return.
This is changing, with the wider adoption of open banking, easy access to cloud solutions, visual solutions like tableau and cutting edge AI and ML packages through python and R. Expect to see a lot more exploring - putting theory into practice.
Two examples we are exploring in more illiquid EM corporate bond markets are:
Machine learning for yield curve feature extraction: Main many fixed income markets, market liquidity is insufficient to facilitate price discovery and as a result, it can be difficult to find bond yields for all rating and tenor pairs. ML can help fill the gap.What causes mean reversion in corporate bond index spreads? The impact of survival around ratings buckets
#FailtoLearnFailToLearn#KeepExploringExperimenting
Distributed Ledger/ Blockchain
The traditional model of financial transactions relies on trusted intermediaries to guarantee the integrity of monetary exchange, which is a slow and costly process. That all changed with the advent of Bitcoin, the digital currency that enabled peer-to-peer monetary exchange without the need for trusted brokers.
Bitcoin itself didn’t gain mainstream traction it initially promised. But Blockchain, the distributed ledger that powers Bitcoin, has risen to prominence in recent years. In a nutshell, Blockchain is a database that is replicated on multiple independent nodes instead of being stored on a centralized system.
By removing intermediaries, Blockchain dramatically reduces the costs and delays associated with completing transactions. This makes it especially convenient for micro-transactions and the exchange of valuables in inconsistent and trustless environments. Among others, the Internet of Things (IoT), supply chain, music and gaming industries are exploring the advantages of Blockchain.
After shunning Bitcoin for years, the finance and banking industry is now embracing Blockchain as an inevitable game-changer. According to Accenture, Blockchain can possibly remove the need for clearing houses and redefine the reconciliation process, resulting in an $8 to 12 billion cut in annual infrastructure costs.
What this means for consumers and businesses is faster and more fluid transactions, lower costs, and wider options for remittance and international payments.
Here are 3 exciting Blockchain applications in Finance
1. Security tokens are cryptographic, programmable securities that serve as an asset that can also take action like pay dividends, interest, etc. - think replacing security certificates.
2. Tokenized assets represent a new form of liquidity in traditionally illiquid assets, such as real estate or art - think new markets where these assets trade 24/365.
3. Self-sovereign identity will allow users to maintain a single digital identity across multiple platforms while selecting the information they wish to share on each - Online logins will be exponentially more secure and efficient. For financial services, KYC and AML work will be transferable from one bank to another, decimating costs.
How Blockchain is likely to disrupt the Bond Market?
At the moment, bonds are issued by companies and banks and subject to a significant number of formalities. The sale of bonds is done mainly through exchanges and brokers.
Key Issues:
In a majority of occasions, bonds are placed on the local market due to legislative restrictions.Mandatory, yet not always justified, involvement of third parties at the stage of issuance.Buyers of foreign bonds lose due to currency conversions.Most bonds are available only for professionals.The absence of a single global platform for issuing bonds.
New Blockchain Model
In the new model, companies, banks and even states can issue obligations in the form of token bonds using Blockchain technological solutions. Bonds are issued as a token on the Blockchain platform. The issuer uses smart contracts as a technological solution for managing security parameters. All transfers are displayed in the public transaction register. The sender and the receiver can be sure that the transfer is complete since they have access to the same data stored in the Blockchain. The management of the income calculation logic and its transfer is done with the help of built-in software solutions - smart contracts.
Blockchain bonds cause the exclusion of intermediaries and change the approach of interaction with investors: everyone, irrespective of their location, can be an investor and can invest in any project in any country.
Advantages
Attracting financing in large volumes and using a more simple procedure than using traditional issues.The company can offer its obligations regardless of the location.Simplicity of involving investors by selling bonds to a wide range of investors. A simple exchange of bonds for money or a cryptocurrency.Open API enables integration with other services: crediting under bonds, circulation on exchanges, exchange of bonds for assets.Absence of exchange rate problems with foreign investors when calculating in national currency. Crypto-exchanges find the optimal exchange ratesRelease does not require permits. Anyone can issue obligations if there is a demand (taking into account the legislative requirements of countries). The proliferation of tokenization within finance will occur, but only expect to see it go mainstream if the problem it's trying to solve, “improving settlement and clearing” is seen as shark bite and not a mosquito bite, otherwise it will take a long time to gain traction.
A few live examples
11 July 2019 - YES BANK implements Asia’s First Commercial Paper Issuance on Blockchain19 June 2019 - EIB, Euroclear, Banco Santander & EY developing Blockchain solution for the issuance and settlement of ECPs15 May 2019 - World Bank and CBA Partner to enable Secondary Bond Trading recorded on Blockchain18 April 2019 - Societe Generale issued the first covered bond as a security token on a public Blockchain11 March 2019 - Deutsche Börse, Swisscom and Sygnum enter into strategic partnership to build a trusted digital asset ecosystem
Open banking
As banking firms try to match the speed and efficiency of the modern connected world, the Open Banking Standard has emerged as a framework for creating, sharing and accessing banking data in a fast, secure and cost effective way.
Open Banking forces established firms to be more competitive with smaller and newer banks. This can result in lower costs, better technology, and better customer service. It also enforces transparent and unbiased publication of information to enable consumers to evaluate the service quality of financial firms. Open Banking APIs allow third party developers to create helpful services and tools that customers can utilize.
The biggest outcome of Open banking is the shift back to customer centric and personalized solutions.
In the future, Fund Managers are unlikely to be able to rely solely on being manufacturers of products, but more likely to be involved in how their products can be more effectively used to achieve clients goals, with fund managers being more flexible than in the past.
Term like Goal-based investing will become more common place with fund managers focusing more on achieving individual targets than purely managing risk.
With new incumbents, like BigTech, very likely to enter the industry, the worst thing investors can do is stay closed to change.
Opening banking will help drive innovation in the industry and investors that look to work alongside, instead of fighting, we see real opportunities.
The clients want more from banks and the client is always right
What is certain is what we do now will not be the same in 5 years.
Appendix
Artificial Intelligence vs. Statistics
AI modeling techniques share many similarities with classic statistical modeling techniques starting from the fact that they both deal with data. However, the key difference, between statistical techniques and AI models is in the goal of these approaches. While statisticians start with a set of known assumptions that are given to the model and best explain the expected behavior of the financial outcome in consideration, AI techniques rather aim at finding by themselves the method (with underlying assumptions that are unknown) that best predicts the outcome in consideration.
Report this
Published by
📷Gareth Nicholson Head of Fixed Income DPM at Bank of Singapore
Comments