Is Amazon a book retailer or a technology firm? Identifying a company’s main business drivers is no longer clear-cut. Martin Small recently sat down with BlackRock Portfolio Manager Jeff Shen to discuss a powerful new classification approach using data science.
Martin Small: Sector allocation has long been used by investors as a tool to implement views on parts of the market that might outperform others. But traditional classification methods tend to group companies using a backward-looking lens. The best-known system, the Global Industry Classification Standard, or GICS, was developed in 1999. Let’s talk about how companies have changed since then and what it means for investors.
Jeff Shen: There are two characteristics to consider here. First, many companies today do more than one thing. Second, companies and their business models evolve as they venture into new areas. As a result, the way we classify these companies also needs to evolve. The obvious example is Amazon, which started as a book seller and over time has ventured into technology with its web services and cloud computing. Now they are adding groceries and even healthcare. For an investor looking to capture Amazon’s growth potential along with other similar companies, what’s the right sector to invest in?
Martin: We know that investors want products that are intuitive and do what they say on the label. That desire for clarity led to the development of BlackRock’s Evolved Sector approach and the suite of iShares exchange traded funds (ETFs). Can you tell me about the technology behind the approach?
Jeff: Yes, what’s different here is that machine learning is at the core of the approach. BlackRock’s systematic active equity team uses proprietary data science techniques, including natural language processing, to dive into how companies describe their business activities and how they talk about the future. Then we use machine learning to identify the cluster of companies that have similar business exposures. It’s intuitive, dynamic, adaptive and uses forward looking inputs. Importantly, unlike GICS our approach allows companies to span multiple sectors. So Amazon is classified both within the technology and discretionary spending sectors.
Martin: How many evolved sectors have you identified?
Jeff: Currently, there are 12 sectors or clusters in the evolved taxonomy. We’re launching first in areas where our methodology makes the biggest difference. For example, in healthcare, where GICS has one classification, we are launching two—healthcare staples and innovative healthcare. We’ve also broken down consumer discretionary into discretionary spending and media entertainment. Our classification updates are done continuously as regulatory filings come in, as opposed to annually. That’s another difference between Evolved Sectors and GICs.
Martin: As an active portfolio manager, most of your day job involves trying to achieve superior risk-adjusted returns for clients. Do these evolved sectors, and the ETFs they’re delivered in, aim to beat the market?
Jeff: Even though I love alpha, these funds aren’t about alpha! They’re about providing investors with a potentially more robust and nuanced toolkit to access sector-level exposure to companies with similar business drivers. Now let me ask you a question. From your perspective, what’s behind iShares’ decision to launch these ETFs?
Martin: iShares’ mission is to provide access to as much of the investable world as possible so that investors have greater control over how they build their portfolios. This means going well beyond the universe of traditional market cap-weighted indexes to offering strategies like factors and scaling technologies that individuals could never access on their own. So while it’s fair to say that most people do not have natural language processing at their disposal, they can now tap into the insights your team gleans from that data. They can do it through one of seven iShares Evolved Sector ETFs, conveniently and at low cost.
See examples of how we use text analysis and machine learning to gain company insights.
Martin Small is the Head of U.S. iShares and a regular contributor to The Blog. Jeff Shen is Co-CIO of Active Equity and Co-Head of Systematic Active Equity (SAE) at BlackRock, and he contributed to this post.