The COVID-19 pandemic has created a new trend line for our society. From changes to our daily routines, to how we pay for things, to how we consume media and stay connected, our lives have been disrupted. Some of these trends were already in place before the pandemic, but now, these changes are likely to be amplified as we transition to a more digitally connected (yet physically separate) world.
A tectonic shift of this magnitude may create tangible opportunities, and investors are likely to seek out specific exposure tools to capture one or all of these next-generation (NextGen) trends. As a result, thematic ETFs are likely to become more popular than those that adhere to the market-cap-weighted archetype or traditional sector frameworks. And over the last few weeks more have been filed, with some launched. But thematic ETFs and their portfolio construction, can vary significantly.
Due diligence is key as these ETF offerings are spread over such an extensive range. The first step in due diligence is to understand what strategies are in scope, and that requires classification. Creating a classification scheme, however, is the windmill that ETF nerds, like me, constantly tilt at. And thematic ETF classification is the mother of all windmills and a true (don) quixotic journey.
Classifying thematic ETFs
Systematic classification is always the goal. In a blog post, I discussed how we created a systematic classification for smart beta strategies. I took the same approach with thematic ETFs, and it almost broke me. I first tried to use Bloomberg’s characteristic data, which is notoriously robust, but failed: it was impossible to systematically classify exposures based on market available data (i.e., market cap, sector, and industry information) AND underlying portfolio traits (i.e., weighting scheme, holdings, etc.). Corner solutions emerged quickly. Funds that should be included were not, and others snuck in – like a traditional technology sector ETF. Every time I tinkered with the code, another problem sprung up. It was whack-a-mole coding.
After hours of trying to engineer a process, I abandoned it in favor of a more qualitative approach that required dissecting all 2,000-plus funds listed in the US. But some quantitative analysis was required, so I culled the list by removing any funds that were easy to exclude (i.e., traditional sector ETFs, as well as broad market cap beta equity, fixed income, and commodity exposures).
To identify the funds, I leveraged applicable portions of New Economies framework developed by our partner S&P Kensho and grouped them into 12 thematic categories based on their fund objective:
- Broad Innovation: Innovation throughout the economy
- Clean Energy: Renewables or firms with low carbon footprints
- Cloud Computing: Cloud storage and cloud based software
- Democratized Banking: Digital payments and encrypted banking technology
- Final Frontiers: Space and deep sea exploration
- Future Communications: 5G networks, streaming media and videogames
- Future Security: Cybersecurity and drone technology
- Human evolution: Advanced medicines and health care solutions
- Intelligent Infrastructure: Smart cities, power grids, and water technology
- New Consumer: e-Commerce and gig economy
- Robotics & AI: Robotics & AI as well as advanced manufacturing
- Smart Mobility: Ride sharing and autonomous vehicles
The result was the identification of 137 funds, comprising $36 billion of assets focused on these particular NextGen trends. As shown below, most of the assets are in funds classified as Broad Innovation and Clean Energy. Clean Energy gets a boost from some legacy ESG ETF positions. However, since the pandemic, there has been noticeable interest over the last few months in Cloud Computing and Future Communications – outside of just broad innovation.