The Role of Inheritance in Assigning Morningstar Quantitative Ratings

When it comes to managed investments, investors have more choice than ever before. Morningstar has been tracking the global growth of the total supply of managed investments—whether exchange-traded funds, open-end funds, separately managed accounts, or model portfolios. By the end of 2021, the number of managed investment products eclipsed 720,000, having grown at a compound annual growth rate of 11.1% for the past 10 years. At this pace, we’d expect to see more than 1 million managed investment products available for sale by 2025 and 2 million by 2032. Indeed, there are some reasons to anticipate that the number of options could grow even faster as the desire to personalize strategies becomes mainstream via the increased adoption by advisors of direct indexing and model portfolios.

Much has been written about the paradox of choice that occurs as choices proliferate. Rather than making us better off, we often end up paralyzed by the sheer amount of options and rarely make an optimal decision in the presence of uncertainty.

At Morningstar, our mission is to help investors make good decisions, and as the number of products has grown, we have also responded by scaling our ratings ecosystem to simplify the decision-making process. Five years ago, Morningstar deployed a machine-learning companion we call the Morningstar Quantitative Rating to sit alongside our Morningstar Analyst Rating and provide investors with an expectation of what our analysts might think about a given managed investment strategy even when they don't cover it directly. By doing so, we've expanded our forward-looking views to an additional 375,000 managed investments from a base universe of 40,000 managed investments covered by our analysts. With this innovation, Morningstar now covers 57% of the available options to investors up from 5%.

Since inception, just like our Analyst Rating,the MQR has been efficacious, stable, and interpretable. Increasingly, we have seen the MQR has become a key component in investor workflows for search and discovery of the best available products. For investors looking to find the best option among a sea of them, the MQR paired with the Analyst Rating serves as an exceptional starting point.

While the methodology behind MQR relies heavily on machine learning to emulate the decision-making of our analysts, we have also invested significantly in our reference data to allow us to inherit analyst judgments across a wider coverage universe directly. Let me give you an example of how this works.

Robby Greengold is our lead analyst on Fidelity Contrafund FCNTX and regularly updates his assessment of the stewardship of Fidelity as an asset manager as expressed by our Parent rating. Given that our goal is to replicate our analyst views on a wider range of managed investment products, we can apply Robby’s Parent assessment to the entire Fidelity fund line-up via a logical mapping. All Fidelity funds should have the same Parent rating. We refer to this as “inheritance,” and we default to it whenever we can. It is only in situations where our analysts do not cover an asset manager that we pull out the machine- learning.

The same principle (albeit with different logic) is also applied to the other two core pillars: Process and People. In these cases, we use different mapping rules and logic relying on a combination of asset class, strategy, and management team.

Over the past five years, we have gotten more sophisticated with how we deploy inheritance logic. Today, over 40% of the ratings issued by MQR inherit all three of their pillar rating decisions. That’s a logical extension of about 157,000 ratings without the use of machine learning at all. Another 51,000 and 71,000 inherit two out of three pillars or one out of three pillars, respectively. Only a fourth, or 91,000 managed investments, rely exclusively on machine learning to score the underlying pillar ratings.

Morningstar believes strongly in a hybrid approach to rating this ever-growing investment universe. It’s simply not possible to cover everything with our team of experts. But through the combination of good data, solid rules, and sophisticated modeling, we are able to help investors continue to make sense of this large menu of managed investment offerings in a way that is consistent with our expert views on what makes a good investment.