In recent years, index providers have launched a variety of thematic strategies to catch the eye of investors with a growing interest in portfolios focusing on niche markets.
MSCI describes thematic investing as “a top-down investment approach that seeks to identify longer-term, structural trends that are expected to be dominant and important explanatory performance factors in a rapidly-changing world.”1 Their thematic indices are designed around “emerging macroeconomic, geopolitical and technological trends that are believed to be structural and transformative.”
Similarly, S&P argues their thematic indices provide investors access to “a series of technologically enabled, often disruptive industries, generally referred to in aggregate as the Fourth Industrial Revolution.”2 Examples of these themes are autonomous vehicles, clean power, cyber security and robotics.
To better understand these new thematic indices, we conducted research to assess their underlying factor exposures by applying the widely accepted asset pricing model of Fama and French.3 Our sample included all MSCI and S&P thematic indices with at least three years of data as at the end of April 2021, giving us 12 and 36 indices from MSCI and S&P respectively.
We observed that almost all these indices have large negative exposures towards the value and profitability factors. On average, the MSCI indices have a value exposure of -0.41 and a profitability exposure of -0.57. For their S&P peers this was even more pronounced, as on average they have an exposure towards the value factor of -0.64 and an exposure towards the profitability factor of -1.18.
In fact, only four out of the 48 thematic indices have a positive exposure towards value, while only one has a (marginally) positive exposure to profitability. These results imply that these thematic indices invest heavily in stocks with weak profitability and expensive valuations, i.e. growth stocks that invest in the present for future profitability.
The MSCI indices have mostly negative exposures towards the investment factor, equal to -0.39 on average. Meanwhile, their S&P counterparts have mixed exposures towards the same factor, with an average of just 0.07. Since the investment factor has been shown to be ‘value-like’ in research,4 this can be interpreted as additional anti-value exposure.
For the momentum factor, the exposures are much smaller and close to zero on average, at 0.07 and -0.10 for the MSCI and S&P indices, respectively. This is not surprising as the momentum factor typically reflects highly dynamic portfolios, while thematic indices are relatively stable. Although a theme may exhibit good momentum at times, there will also be periods when it is out of favor. As a result, we would not expect to find large momentum exposures over the long run.
To put this into perspective, we compared the factor exposures of these thematic indices with those of various other common portfolios. Figure 1 shows the aggregated exposures towards the value, profitability, investment and momentum factors for the S&P and MSCI thematic indices, alongside 11 global GICS sector indices, 49 US industry portfolios, US decile portfolios for stocks sorted on value, profitability, or investment, and US portfolios double-sorted on value and profitability, value and investment, as well as profitability and investment. Each dot in the graph represents an individual portfolio.
Figure 1 | Aggregated exposure to the Value, Profitability, Investment and Momentum factors of thematic indices versus various other stock groupings, June 2013 – April 2021
Source: Robeco Quantitative Research
It is clearly visible in Figure 1 that virtually all the MSCI and S&P thematic indices have negative aggregated factor exposures and that their magnitude is sizable. Effectively, investors in those indices are trading against quantitative investors who seek positive exposures to the same factors. From an asset-pricing perspective, this implies that investors in passive thematic indices could potentially face lower long-term expected returns. But there are a number of reasons that could explain their rationale for betting on these indices.
Firstly, these investors might expect factor premiums to be zero or perhaps even negative in the future, due to changes in investor sentiment or macroeconomic conditions. For instance, academic research finds that factors can experience entirely lost decades.5 More specifically, investors in thematic indices may believe that we are in a period of major industrial transformation, which favors growth stocks over their value peers.
Secondly, investors in thematic indices might believe these ETFs can deliver solid outperformance, in spite of their negative factor exposures, due to the idiosyncratic dimension. Although the average stock with weak profitability and expensive valuations may underperform, specific themes may still be successful despite sharing these characteristics. Related to this, investors may be attracted to thematic indices because of a preference for lottery-like payoffs.6
Thirdly, these investors may be attracted to thematic indices because of historical returns. Researchers found that ETF providers launched funds based on niche themes with strong past performance, high media exposure and positive sentiment.7 However, they also found that such funds tended to perform poorly after their launch, as the hype around them vanished.
A fourth and final explanation could be that investors in thematic indices are willing to accept a lower financial expected return due to unfavorable factor characteristics because they also take non-financial considerations into account. This argument applies in particular to sustainability-related themes, such as clean energy.
Altogether we conclude that various arguments can rationalize the choice to invest in thematic indices, despite their unfavorable factor exposures. Moreover, if a theme has negative factor exposures, investors may consider screening out the stocks with the worst characteristics, for example low profitability and expensive valuations. In other words, active management might be applied to improve the factor exposures of a thematic investment strategy.
1See: MSCI, November 2019, “Indexing change: understanding MSCI thematic indexes.”, MSCI website.
2See: S&P, September 2021, “S&P Kensho indices methodology.”, S&P website.
3See: Fama, E.F., and French, K.R., April 2015, “A five-factor asset pricing model.”, Journal of Financial Economics 116 (1): 1-22.
4See: Fama, E.F., and French, K.R., April 2015, “A five-factor asset pricing model.”, Journal of Financial Economics 116 (1): 1-22; and Blitz, D., and Hanauer, M.X., May 2021, “Resurrecting the value premium.”, Journal of Portfolio Management 47 (2): 63-81.
5Blitz, D., June 2020, “Factor performance 2010-2019: a lost decade?”, Journal of Index Investing 11 (2): 57-65.
6Shefrin, H., and Statman, M., June 2000, “Behavioral portfolio theory.”, Journal of Financial and Quantitative Analysis 35 (2): 127-151.
7Ben-David, I., Franzoni, F.A., Kim, B., and Moussawi, R., July 2021, “Competition for attention in the ETF space.”, SSRN working paper, no. 3765063.