PE Tech Report


Like this article?

Sign up to our free newsletter

The impact of AI and machine learning on the hedge fund industry

Data re-use is likely going to become a critical feature of the industry, moving forward. Not only will fund managers want to be able to access data in a consistent fashion, they will also want to make sure it is accurate across different reports as regulators pay ever closer attention. 

Rather than spend numerous man hours organising data to complete mundane tasks, the funds industry stands on the cusp of an exciting new era; one that will see increased use of machine learning and AI technology taking over such tasks, freeing up managers to hone investment strategies, and build investment solutions that more closely match investors’ needs, based on understanding their behaviour.

Everywhere one looks, blockchain technologies are sprouting up. Goldman Sachs, for example, has filed patents for various blockchain technologies including SETLcoin. Another example of this is CORDA, a bitcoin variant developed by R3, a consortium of banks that are pooling resources to research the potential of blockchain.

Indeed, the immutability feature of blockchain is what makes the ability to look at information and perform reconciliations in real-time, for example, really exciting. It has the potential to revolutionise the financial industry by removing reliance on intermediaries to clear and settle accounts and introduce significant cost savings. One survey1 by Bain & Co, estimates that total savings to global financial markets could reach anywhere from USD15 billion to USD35 billion.

On 28 June, RBC Global Asset Management (RBC GAM) announced the establishment of a new RBC GAM Innovation Lab, an in-house technology hub to incubate digital capabilities and drive innovation. 

Leveraging the network of technology and innovation capabilities across RBC, the RBC GAM Innovation Lab will focus on developing and executing next-generation initiatives to enhance the experience and outcomes of investors and advisors.

“The culture of RBC Global Asset Management revolves around innovation, continual learning and harnessing the power of human and machine,” said Damon Williams, CEO, RBC GAM. “Our new Innovation Lab is a reflection of this philosophy.”

RBC GAM has committed initial funding of USD20 million over the next five years to establish the Innovation Lab and fund its initiatives.

Ajantha Ganeshalingam heads up RBC GAM’s Innovation Lab. He says that the team is looking at various capabilities from portfolio creation through to digital communication between portfolio managers, advisors and clients. “Then there are the analytics capabilities, which is an area we are actively trying to grow; generating insights, understanding customer behaviour and much more. 

“In terms of AI and machine learning playing a role, we want to be able to plug in behavioural finance as an input to assist a client in their selection process. Rather than asking someone what their risk appetite is, with respect to investing, how might we use insights about a client’s behaviour such as spending habits or what they like and talk about online to help them evaluate their risk profile and make more informed financial decisions,” says Ganeshalingam. 

The Innovation Lab’s strength is that it has a helicopter view on where it wants to go, whether be to develop internal machine learning solutions or develop more niche areas with respect to fund products. 

It is only in the early stages of road mapping what that end state could look like. 

“The exciting part over the coming months will be looking at what are all the different capabilities currently being built and researched within RBC’s ecosystem and to see what we can leverage in the investment space. Can we provide portfolio managers or product owners with better insights on clients, and comparisons with other similar fund products in the marketplace to influence RBC’s decision-making process? 

 “In financial services the balance of power is shifting. Information and insights on financial products has always been kept in the hands of the financial firms but I think that is slowly shifting. Clients are now being provided with much more data, presented in a simple manner,” says Ganeshalingam.

To see how much interest there is, in relation to machine learning, one only has to Google the number of innovation labs that are emerging. Earlier this year, for example, Deloitte established its own dedicated blockchain lab in Dublin. This will be its base for the creation of an EMEA Financial Services Blockchain Lab as part of its FinTech initiative, `The Grid’. A 50-person team will focus on developing strategic blockchain capabilities and proof-of-concepts into functioning prototypes to create `ready to integrate’ solutions for financial services clients. 

“As a firm we have invested a vast amount of effort in reviewing exponential trends and technologies; more specifically, we started building our blockchain development capability about three years ago. We have made a significant investment in our belief that blockchain will be a critical enabler and core technology for the future,” explains Cillian Leonowicz, Senior Manager Consulting and Deloitte and the Ireland Firms FinTech Lead.

Deloitte will take a two-pronged approach: on the one hand it will build provide bespoke services from strategy development to production blockchains for individual clients, on the other hand it will build industry solutions and platforms to serve as a future fabric for industry and in particular financial services.

“This is where we see the organisation going in terms of innovation and building a new core within a traditional business. The lab is a new interactive space, which is quite different to our traditional consulting environment. We focus on two things. 

“Firstly, services consulting, where we apply traditional consulting practices and our digital experience to blockchain. We can do everything from strategy development and ideation right through to full end development. We will have our first live production client – a Nordic-based client – on the 2nd September 2017.

“Secondly, we focus on what we call solutions or `bold plays’. These are platform capabilities that we will build with selected partners, primarily from the financial services world, to build the future fabric of the industry. We are looking at a number of things in this space. One is a KYC utility, a second is regulatory reporting services using blockchain technology.

“We’ve already built a solution we call `RegChain’, this ingests fund data from core systems, runs smart contracts within the blockchain to meet the requirements of regulatory reporting, creates the reports and provides a node to the regulator for review,” explains Leonowicz.

Leonowicz thinks that in order to change the future fabric of the funds industry, three things need to happen: 

  • To get customers on the blockchain platform via a KYC or identity capability and a fund wallet; 
  • Tokenise funds or shares and allow for them to be traded on the platform; and 
  • To facilitate the regulator and determine how they want to interact and serve in this new paradigm. 

“Beyond this we will need to provide the governance frameworks and ensure sustainability as well as scalability of the platform. Once you have these three pieces in place, you’re close to having the key elements needed for replacing the traditional structure of the fund industry potentially without the existing roles of custodians and distributors. What we believe is that in future, when all shares and bonds are tokenised onto blockchain we could get to a stage where everything is priced on the blockchain in real time,” suggests Leonowicz.

Of course, this technological revolution has not been lost on fund managers. One high-profile example is New York-based Protégé Partners, one of the industry’s leading seed investors. As part of its roadmap to look for tomorrow’s future talent, Protégé established a separate entity named MOV37, whose modus operandi is to invest in what it calls Autonomous Learning Investment Strategies (ALIS). 

Jeffrey Tarrant, CEO and Founder of MOV37 and CEO, CIO at Protégé Partners, explains how he arrived at the Eureka! moment: “A couple of years ago when I was at a Ted Talk in Vancouver, I heard Jeremy Howard speak on the future of Artificial Intelligence and how computers were already on the cusp of surpassing human performance in many tasks. Shortly thereafter, I met a fund that was modelled after Graham and Dodd’s value investing philosophy but was able to analyse every public filing, such as 10Qs and 10Ks since inception. This further confirmed my belief that this was the future of investing.”

In a fascinating white paper – The Intelligent Investor in an Era of Autonomous Learning2 – Tarrant details the rapid arc of AI capabilities, which culminated last year when AlphaGo, an AI system developed by Google DeepMind, beat Lee Sedol, the world’s best Go player, demonstrating an incredible ability to not only think like a human being, but surpass one.

The idea behind ALIS is essentially that the coming together of man, machine and data science is set to create the `third wave’ of investment management. 

The first wave were fundamental discretionary investors, says Tarrant. The second wave were computational science, systematic quantitative managers who used hypothesis-driven programming and structured financial data. 

“The third wave are ALIS managers who use machine learning, which is typically data (rather than hypothesis) driven and unstructured and non-financial data (rather than structured financial data). 

“The first wave was run by MBAs, the second wave was run by PhDs in finance and the third wave is run by PhDs in science, machine learning, robotics, computer science and other similar fields.”

Michael Weinberg is Chief Investment Officer (CIO) at Protégé Partners and a Partner at MOV37. He says that the key reason investors should be excited about ALIS managers, is because they are for the first time “viable now”. 

“It is a confluence of five factors that makes these strategies possible and these factors didn’t exist previously. First, the massive growth in unstructured data. Second, the field of data science. Third, the record decline in data storage and processing costs. Fourth, machine learning is working better than it ever has over prior decades. And fifth, we are in an era that is not conducive to fundamental managers getting better information. Now, systematic strategies have an advantage to fundamental, discretionary ones.”

“Our team has spent the last two years going around the world meeting with approximately 135 ALIS managers. For context, one of the preeminent database providers only has approximately 25 such managers in its database,” outlines Weinberg. 

This is just one example of how the funds industry might start to embrace AI systems to create centaurs; humans and machines joining forces to produce the third way of investing referred to above. 

The best managers have robust systems that are in fact learning, using unstructured non-financial data, and derive their profit over multiple securities, market regimes, sectors and on both the long and short-side. 

“We do not tend to invest based on back-tests, hypotheticals or pro-formas. These managers also generate returns that are alpha, not beta, driven and have little to no, or even negative, correlation to the hedge fund and long-only indices,” adds Weinberg.

There are concerns, perhaps legitimately, that AI systems, left unchecked, could use their powers for malign reasons. In a portfolio management context, one example often cited by Elon Musk is that an AI system could go long defensive stocks, short consumer stocks, and initiate a global conflict to maximise profits; there would be no human empathy involved. 

Ganeshalingam concedes that AI will cause human roles to changes but not too drastically. People will be equipped with better tools with which to make better decisions, whilst achieving operational efficiencies along the way. 

“It’s a complementary service, at least in the near term. In the long term, it will depend how the technology evolves. 

“At the RBC GAM Innovation Lab, we are more focused on applied innovation i.e. looking at existing capabilities within the business, and working out how to better position RBC to be a better positioned with clients in the marketplace. 

“We want to spend time understanding the capabilities of existing technologies, and design a strategy around applying the internal AI research in parallel to understanding the challenges in the investment management space. The short of it is RBC is investing significant resources and capital to be a leader in the AI and machine learning space,” concludes Ganeshalingam. 




Like this article? Sign up to our free newsletter