Adam Riley, head of international wealth platform strategy for BlackRock Systematic, talks to Piyasi Mitra about how the next generation of active ETFs brings together big data and active investment insights.
BlackRock recently launched five European-listed equity-enhanced active ETFs that provide investors with access to low-cost, asset allocation building blocks that offer consistent alpha-generating potential at the core of their portfolio. The range combines BlackRock’s active management expertise with the breadth and scale of iShares, the ETF provider powered by BlackRock.
The new funds are devised to outperform their parent benchmarks and target consistent, repeatable alpha with low tracking error. The investment process gives access to small, evidence-based positions across a universe of securities in real time, thanks to the combination of human expertise and big data, which remains a competitive edge.
Here, Adam Riley, head of international wealth platform strategy for BlackRock Systematic, walks through the team’s disciplined portfolio construction techniques focussed on delivering differentiated investment outcomes.
Q: With the launch of the new active ETFs, how does your team use big data analytics to optimise performance in the active investment process?
The BlackRock Systematic platform manages $220 billion in assets and has been managing client investments since 1985. Our investment approach emphasises data-driven insights and the scientific testing of investment ideas, with the team’s portfolio construction techniques focused on delivering differentiated investment outcomes.
As part of our culture of innovation, we started investing in data science talent in 2007, with our first projects focusing on alternative big data and natural language processing based on analyst and broker reports. Applying text analysis algorithms to analyst reports can provide better insight into whether analysts have positive or negative views of the firm. Additional measures of value can also be captured by alternative data signals, such as the number of web searches for a specific company’s products.
This evolution has taken a significant investment in talent, systems and data over the past 17 years. Today, half of our team of investors, researchers, and strategists have academic backgrounds outside of finance, including data science and engineering.
“Rather than viewing the challenge of combining signals as a weight allocation problem, we reframe it as a forecasting problem. This combination of human expertise and machine learning makes our approach powerful for constructing systematic portfolios while balancing risk and return considerations.”
Q: How does alternative data affect your portfolio management process and decision-making?
The BlackRock Systematic equity team conducts over 100 data trials annually and has developed technology that enables running a five-year back-test in just over a second to test these new data sets. In recent years, we have developed large language transformer models, and we apply these to text across news, earnings call transcripts, job postings, social media, financial statements and broker reports.
To understand the drivers of stock returns, we research investment insights related to company fundamentals, market sentiment (technical), the macro environment, and ESG, and we use alternative big data across each of these research areas.
Q: How do AI and machine-learning contribute to BlackRock Systematic’s investment strategies?
Advances in machine-learning offer powerful tools to help address the challenges faced by investment managers in designing alpha models. Machine-learning algorithms, such as decision trees or deep neural networks, have demonstrated remarkable capabilities in extracting patterns and relationships from complex, large-scale data sets. For this reason, machine learning tools offer a dynamic alternative to designing alpha models through a discretionary approach.
Rather than viewing the challenge of combining signals as a weight allocation problem, we reframe it as a forecasting problem. This approach requires combining signals into a forecast of excess returns, which we call alpha forecasts for each asset in the investible universe.
Starting in 2014, we set out to design a system that can train machine-learning models. Today, we combine our systematic investment models looking at fundamentals, sentiment, macro and ESG investment insights using this system to build diversified portfolios supported by a modern investment approach. This combination of human expertise and machine learning makes our approach powerful for constructing systematic portfolios while balancing risk and return considerations.
Q: Describe the technologies and platforms your firm uses to enhance portfolio management.
The investment process encompasses multiple key parameters: forecast returns (expected alpha), risk (tracking error), ESG data (MSCI ESG scores, carbon intensity) and transaction costs.
The decision to buy/sell a stock is made in the context of these key parameters, given the investible universe as well as portfolio constraints and exclusions.
Portfolio managers are responsible for the oversight of the investment process. This includes assuring the integrity of inputs and reviewing recommended trade lists and portfolios to ensure that the process is functioning as intended.
The final portfolio for any fund consists of small active holdings, balancing diversification and risk control while aiming to deliver returns over the benchmark. Portfolio managers review and optimise the fund regularly to account for new information derived from the investment model.
As the world’s largest asset manager, BlackRock’s scale uniquely positions our clients to access liquidity and secure the most competitive pricing from top broker-dealers.
“Despite the advances in modelling and growth of computing power, the role of human talent has never been more important, so it is key we continue to hire the best talent in the field to research new investment ideas and implement these in portfolios.”
Q: What are the challenges and opportunities of incorporating technology and big data into the active ETF space?
A key challenge with the advancements of technology and the open-source nature of large language models such as ChatGPT is to ensure we stay ahead of the competition in terms of new investment ideas and talent acquisition.
Despite the advances in modelling and growth of computing power, the role of human talent has never been more important, so it is key we continue to hire the best talent in the field to research new investment ideas and implement these in portfolios.
We are facing a new market regime of greater macro and market volatility, where investors will need to take a more dynamic approach with both index and active strategies to achieve portfolio outcomes. Many investors are increasingly looking for differentiated sources of return and want to access these themes through the wrapper of their choice.
