The Seven Sins of Thematic Investing

Long-term investors are highly susceptible to what we call the narrative fallacy - an appealing story that fails to translate into long-term investment returns. We believe a theme should represent not just a broad idea but a potentially positive long-term investment opportunity. The world of thematic investing is unfortunately rife with products that capture the imagination but on further scrutiny do not confer a genuine benefit in terms of return, risk, or sustainability objectives. This commercial window dressing does a disservice to more robust thematic strategies, which may find themselves dismissed without due consideration.

The narrative fallacy problem poses a particularly potent risk when a single theme is the entire investment proposition. Single-theme strategies should come with a crucial acknowledgment: They outsource to the client one of the most critical investment questions - is this theme a good investment?

To be clear, single-theme strategies can be robust investments, and some clients and asset owners will have the knowledge and expertise to select winning themes. Yet, single-theme strategies compound agency risks around confirmation bias, in which the portfolio manager seeks corroborating evidence to support the thematic thesis and discards evidence that undermines it. The alternative - retiring the theme and hence closing the strategy - is typically an unpalatable option.

In contrast, a key benefit of multi-theme portfolios is that they ensure competition for capital across themes. When the retirement of a theme is not an existential crisis for a money manager, but rather an allocation decision, it can be approached dispassionately. There is every incentive to identify the themes that are truly compelling at any given time and no penalty for an honest assessment that a theme has run its course.

The source of thematic ideas matters a lot. Popular perceptions of the future are frequently incorrect in terms of magnitude, direction, or timing. The primary aim of both the media and many “thought leaders” is to attract eyeballs, rather than to offer up rigorous accountability, which should be an unsettling thought to those investing in products that treat these sources as primary proof points and actionable information.

We believe that thematic insights should be grounded in real-world experience. That’s why company management teams are the primary source of our understanding of the most significant structural changes of the next decade. We meet with a large number of management teams each year who offer direct data points about structural changes within their industries, economies, and markets. As the companies will likely be key enablers of these structural changes and are deploying capital to make them happen, their views represent an invaluable primary source.

It should be noted that we approach our conversations with a healthy dose of skepticism. We supplement them with information from our global research team and seek conflicting views by speaking to, for example, both incumbents and disruptors. Having the experience and expertise to properly evaluate and weigh the information that comes out of these discussions is, in our view, a widely underappreciated investment attribute. We discuss this further in Sin No. 7, “The Wrong Resume”.

Insights from our meetings form the basis of our Global Framework, an overview of how we think the world will change over the next decade. Yet, crucially, structural changes do not happen in isolation. The Global Framework blends together all the key structural changes we identify in a single place, ensuring we consider the interactions between them. This is preferable to the unrealistic assumption that any strategy can perfectly isolate a single structural change.

Even if they are inspired by a manager’s vision, investors should also expect the people managing their money to be humble about their ability to know the future. For thematic managers, this humility needs to be incorporated from first principles, starting with the investment philosophy and carrying right through to stock selection and position sizing. Features such as 10-year target prices, extreme outsized positions in single stocks, or excessively precise optimization of theme or stock weights are, in our view, red flags that suggest hubris.

Many “thematic” managers invest in generic, easily recognizable themes that simplify the marketing process. Yet the narrative fallacies of Sin No. 1 and the humbling experience of forecasting elucidated in Sin No. 2 suggest that we should be deeply wary of assumed knowledge. Extending this thought process to the world of standardized, commonly accepted, generic themes raises a deeply uncomfortable question for many thematic managers: If an investment theme is truly inevitable and widely accepted as such, surely it should be already fully discounted?

We believe that companies that are obvious beneficiaries of popular narratives are often fully valued. Our approach therefore rests on the creation of proprietary themes, each of which attempts to capture a number of structural changes that are likely to result in long-term value creation over and above the common consensus embedded within generic investment themes. In creating these proprietary themes, the scope of each theme warrants careful scrutiny. In particular, we believe that themes are often defined too broadly, necessitating a number of sub-steps to reach the underlying investments. Each of these steps layers assumption on top of assumption. Theme design is an art, not a science, but investors should expect to hear some explanation as to how the thematic insights might play out in terms of fundamental outcomes. We prefer the precision of a scalpel, isolating the intersection of structural changes and investment return opportunities, over the sledgehammer of a grandiose vision of the future.

Many structural shifts follow adoption curves driven by the disruptive impact of technology, but we do not believe early-stage disruption is the only area of relevance for thematic investing. Investing in the early stages of disruptive cycles can offer significant opportunities due to the non-linear pace of change, but we must acknowledge the risks as well. Products or technologies can fizzle out after initial excitement, and many disruptive companies have been overwhelmed by fast followers or deep-pocketed incumbents.

We often see an improved level of risk and reward a little later in an adoption curve, as product offerings go mainstream, industries consolidate around one or more leaders, and the potential for the emergence of stable profit pools becomes more apparent.

Similarly, we do not believe that a thematic approach should be limited to a growth style. There are many forms of structural change, as detailed in our companion paper, “Capturing Structural Change.” We consider companies in all stages of their life cycles when constructing themes.

Finally, even good themes play out over time, and investors who aren’t careful about knowing when to move on may find themselves desperately trying to justify forecasts and valuations. A more disciplined approach would insist that themes must evolve over time or face retirement. Consistent with our scalpel comments above, our themes may become more precise over time as we gain new insights. We believe a dynamic approach, while a little more complicated to explain than a static approach, best represents the reality that the world changes over time and we must stay ahead of these changes and market consensus rather than assume comfortable constants.

The expression of themes through stock holdings requires thorough scrutiny. This is because stocks that appear to be valid candidates for a theme might actually have very little relevance. Whether due to lack of rigor or sleight of hand, investors in a theme may not be getting what they signed up for.

Some managers refer to the concept of a stock’s thematic fit as “purity” and quantify that purity by mapping to a simplistic investment driver such as the current or future revenue mix. We do not consider that kind of analysis sufficient for determining genuine thematic fit, however. Broad revenue exposure to a thematic narrative is meaningless without understanding the specific structural drivers involved. We believe identifying stocks that can benefit from the precise structural changes behind a theme, not just those that have some weak form of exposure to them, is the only way to determine true thematic fit. In addition, a simple focus on revenue exposure suggests a focus on companies selling products that are exposed to a shift in demand. However, product cycles are notoriously hard to forecast, particularly given the pace of change in disruptive areas. Many products have generated high short-term sales but low long-term profitability. We generally consider such product-driven investment themes as high risk. A product cycle is not a theme.

We take a broader view of structural change than the size of an addressable market or revenue growth, as a thematic thesis could also play out as a wider moat, a longer duration for returns, or the potential emergence of new opportunities. Ultimately, we anticipate these outcomes to be reflected in improved long-term fundamental performance, which includes not only revenue but also margins, cash generation, returns, and cyclicality, to name but a few. A broad, inclusive view of how structural change can manifest itself produces a better sense of thematic fit.

Stocks that fit the thematic thesis will have additional, idiosyncratic factors that drive future performance. Stock-level analysis should at a minimum ensure that these factors do not materially offset the thematic merits of the investment. Ideally, a stock’s idiosyncratic investment considerations would actually be additive to overall investment asymmetry. In either case, access to a deep, independent research platform is an essential cross-check.

We often hear that small or mid cap companies are better thematic candidates because they are “purer plays.” We have some sympathy for this point of view in certain cases, but we are loath to disregard decades of study into the significant benefits that accrue to large companies because of some arbitrary measure of “thematic purity.” On the other hand, if we only use large cap companies to populate themes, we risk losing focus on the thematic fit and access to smaller, more disruptive companies. We believe drawing from companies of all sizes is a reasonable compromise to optimize the theme between thematic fit and idiosyncratic factors such as incumbency advantages and economies of scale.

Finally, there is the question of valuation: Our philosophy is to focus on key inputs that matter to long-term valuations. Given that the majority of a stock valuation is driven by its prospects 3– 10 years from today and beyond, our emphasis is on changes to these expectations. We believe the presence of structural change, as reflected in our theme design and stock selection, is a strong foundation for differentiated long-term performance. Our work on valuation allows us to assess whether our insights are material and whether they are reflected in the stock price today.

Anchoring to themes instead of a benchmark poses a range of questions around portfolio construction, particularly risk management. Traditional benchmark-centric approaches typically target relative portfolio characteristics such as active share and risk metrics such as tracking error, yet clearly these are not relevant to a benchmark-agnostic strategy. So how should a thematic portfolio consider risk?

In Sin No. 1, we discussed the relative risks of strategies based on a single theme as part of a larger concern with narrative fallacies. The cousin of the single-theme strategy is the “thematic” fund that is really a single-factor fund. In our view, a thematic fund should try to access multiple sources of return from structural change without permanently embedding a particular factor based on style, sector, or geography. To the extent that a single dominant factor cannot be mitigated or is part of the overall process and portfolio construction, it should be openly acknowledged and disclosed, in our view. Investors need to ask themselves if, under the guise of a “theme” they are being asked to pay active management fees for access to a relatively simple factor.

Instead, our approach returns to first principles. We are investing in long-term structural change, which is uncertain. The best defense against this uncertainty is to ensure that we are anchored to many different structural drivers, and not just putting all of our eggs in one basket. We have already established that we can capture exposure to a number of structural drivers in a single theme. By combining a number of themes together we can further enhance potential diversification benefits, provided that the themes are not driven by the same aspects of structural change. So, the first task is to ensure that themes are genuinely different from each other. We believe that this diversification at the theme level is superior to geographic or sector-level diversification, as it is driven by fundamental analysis of what actually drives businesses rather than arbitrary benchmark classifications.

How should capital be allocated between themes? Clearly, we should wish to optimize for perceived long-term return and risk, yet as we saw in Sin No. 2, “Foggy Forecasting”, we need to be humble about our ability to predict future outcomes amid unforeseeable risks. We typically try to design themes so that the overall perceived asymmetry - considering both long-term opportunities and downside risks - is broadly similar for each theme. For this reason, our portfolio generally consists of a number of approximately equally sized themes. Only if our qualitative and quantitative analyses suggest that a theme could potentially dominate overall portfolio performance at an equal weight would we reduce the weight of that theme to attempt to normalize its contribution to the portfolio.

Similarly, we believe all stocks in a theme should contribute equally to the theme’s return and risk profile, which normally translates into broadly equal position sizes. Occasionally, we may hold a security in a substandard position size, either because the company has high levels of idiosyncratic risk or because we believe certain risks are best mitigated through additional diversification. In the latter circumstance, we typically use a basket approach to implement the idea.

Quantitative tools can assist in risk management at both the theme and the stock levels. For example, the team monitors inter-theme correlations, theme volatilities, and overall portfolio volatility data as prompts for revisiting fundamental views. Though our strategies are benchmark-agnostic, we may monitor (but not manage) common benchmark-centric risk metrics to help us understand portfolio exposures. Yet, we also know that over the long run correlations and volatility metrics can change. Using both qualitative and quantitative processes are the best way to understand and manage risk.

In summary, we seek to mitigate exposure to unintended risks at the portfolio, theme, and stock levels. Managers need to be aware of dominant factor exposures and disclose them. It is also easy to become enamored with a theme or stock and allow either to dominate returns. Fundamentally different themes, implemented via broadly equal risk-adjusted weights at both the theme and the stock levels, can take the emotion out of portfolio construction decisions and acknowledge that there is much we cannot know.

When initially identifying structural change, many thematic strategies fail to incorporate an analysis of non-financial externalities into their view of the future. A broader approach acknowledges that insights may focus on either traditional industry change (e.g., competitive dynamics, industry structure, disruptive technology) or a wider range of ESG and sustainability-related issues including cultural shifts, changes in societal attitudes and norms, and shifts in regulations and policy.

Unfortunately, greenwashing - the largely cosmetic bolstering of a product’s sustainability credentials, primarily for marketing purposes - is widespread in the investment industry. This is as true in the thematic equity space as it is in other areas. The robust alternative to greenwashing is to fully integrate sustainability considerations into all stages of the process - identification of structural change, theme creation, and stock selection - and to think as hard about non-financial change as we do about more traditional areas of financial analysis.

Our Global Framework incorporates a holistic range of inputs, including those which suggest long-term shifts in societal norms. For example, we consider the direction of future policy through ongoing shifts in existing policies on global issues such as climate change. As policy frameworks evolve, our Global Framework can evolve alongside them, helping to ensure our view of the world is not locked to a redundant or obsolete policy model.

The analysis of the likely direction of future policy can yield valuable structural investment insights - we want our themes to be on the right side of future policy change. Specific sustainability goals such as the UN SDGs may be useful indicators of future policy direction. Whether specific themes “target” these goals explicitly or the theme construction process incorporates an analysis of alignment with these potential policy shifts is largely, in our view, semantics - again, we note that specific policy frameworks do change over time and hence can become redundant or obsolete. We can map the positive alignments between our themes and, for example, climate change policies or UN SDG solution pathways to demonstrate where we believe our consideration of sustainability goals provides additional support for our investment views. In all cases, the aim should be to enhance returns by integrating both financial and non-financial structural shifts into theme construction, such that returns may be enhanced.

At the stock level, we would highlight two further considerations that we believe can enhance returns, and mitigate sustainability-related risks.

First, all themes are aligned with what we see as probable policy change. Since we insist that all stocks are chosen specifically to fit a particular theme, all stocks in a portfolio are likely to be on the right side of forthcoming policy changes. Thus, we believe the stocks we invest in are part of the solution to societal goals, rather than part of the problem.

Second, all stocks considered for the portfolio are subject to our Sustainability Framework which integrates multiple aspects of business risk, including formal ESG inputs. This framework focuses on the strength of the relationships companies and industries have with society - the societal license to operate - and how this might change over time. We believe the breakdown of these relationships is the key mechanism through which sustainability risks impacting company fundamentals. The Sustainability Framework consists of a three-step process that assigns companies a series of scores, each incorporating a direction of change, which must pass a minimum threshold level to qualify them as potential investments. Our Sustainability Framework aims to ensure that companies exposed to the risk of a deterioration in societal license are excluded from the portfolio. More details are available in a separate paper².

The final category of potential mistakes revolves around the organization and expertise of the portfolio team and the broader resources of the asset management firm.

We believe strongly that experience in analyzing structural change is very valuable. An investment manager need not be characterized as a “thematic manager” to know plenty about structural change. We do, however, think the inverse is true. A deep, broad knowledge of structural change should be a prerequisite for a manager purporting to have thematic investing expertise. Indeed, a client appointing a thematic manager is probably doing so in the belief that the manager specializes in understanding structural change and how to translate it into investment objectives. Yet, learning about structural change can be a slow process and is often assumed to be only inherited by osmosis after years in the market.

In our experience, covering a specific geography or industry, even for decades, might not produce enough learning opportunities to provide the appropriate context when analyzing structural change. Structural change tends to cut across industries, styles, and geography, blurring the boundaries between them and creating a high degree of commonality among the challenges facing companies today. There are therefore good reasons for thinking about structural change in unconstrained terms.

We feel that a global cross-sector approach offers the greatest opportunity to learn how structural change really occurs. We have typically trained our own team members in the key concepts and constructs behind the analysis of structural change and the implementation of a thematic strategy. Over time, this knowledge represents an information advantage that matters more than the investment team’s years of experience.

In terms of skill sets, thematic investing requires a blend of lateral “big-picture” thinking and attention to detail which takes time to accrue at both the individual and the team level. We also value a collegial approach, as the ability to debate openly every possible aspect of an investment decision in a constructive environment is of great cultural importance. No one person has all the answers.

Investors should also consider how a portfolio management team uses the broader resources of the firm - specifically, the research analysts who have crucial knowledge of sector and idiosyncratic stock issues. Is the optimal research team composed of analysts dedicated to a specific theme, or should the team leverage a firm-wide research platform? Each approach has advantages and disadvantages, but we feel the alignment of interests is the most important factor. Analysts dedicated to a particular theme face the same problem as managers of single-theme strategies: confirmation bias. They may be tempted to ignore or discount evidence that the theme is no longer relevant because acknowledging it is arguing for their own obsolescence.

We resolve this tension by working closely with the firm’s broader research team but ultimately maintaining a respectful independence. Our research analysts support a broad range of mandates, including ours, and we can leverage their expertise and insights from over 4,000 company interactions per year. We view the presence of an independent research platform as a competitive advantage.