Key Takeaways

  • Morningstar’s original moat framework wasn’t built on theory. Pat Dorsey explains they analyzed historical data from companies that consistently sustained over 15% returns on capital for 15+ years.
  • This data-driven approach revealed four recurring moat categories: intangible assets (like patents or brands), high customer switching costs, network effects, and scale advantages (cost advantages).
  • Dorsey now argues that Return on Invested Capital (ROIC) is far less useful for identifying moats in modern, capital-light software and service companies, where the 'denominator is nothing.'
  • For today’s businesses, identifying a moat requires a predominantly qualitative lens, focusing on true pricing power, scale economies shared, and foundational attributes that defy economic gravity.
  • The Morningstar's Original Moat Framework Development offers a robust, data-backed method for dissecting how enduring competitive advantages were historically built.

The Morningstar's Original Moat Framework Development

Pat Dorsey recounts how Morningstar approached the seemingly abstract concept of a competitive moat: they went straight to the numbers. The goal was to identify what traits businesses with sustainable, superior returns actually possessed.

Initial Quantitative Screen: Looked at every company that had done more than 15% returns on capital for more than 15 years. Totally arbitrary numbers, but the idea was basically, instead of theorizing, let's just look at the data.

Pattern Observation: Go and get the companies that have done this and generated sustainably high returns on capital and see if we can observe patterns.

Moat Categories Identified: Most of the companies that had done that, you know, could be sourced to some kind of a intangible asset, like a brand or a patent or a government approval, you know, high customer switching costs, like you see with databases, network effects, or scale advantages, cost cost advantages. And you know, most of them kind of fit in one of those buckets.

Framework Adoption: Well, that's what the data says. Let's Let's use that framework going forward.

When This Works (and When It Doesn't)

This framework was initially developed using data from the 1960s through the mid-1990s. This was an era before software companies dominated the market, and capital was generally a more significant factor in business operations. It excels at identifying the foundational attributes that allowed traditional, often asset-heavy, companies to sustain high returns on capital despite competitive pressures. If you're looking at manufacturing, retail, or other industries with tangible assets, this framework offers a solid historical lens.

However, Dorsey himself acknowledges its limitations today. He notes, “Today, I would argue actually return on capital is much less useful as a touchstone for does a company have competitive advantage because, frankly, if you don't have any capital, it's pretty and your denominator is nothing, it's pretty easy to generate a high high ratio, right? It's just math.” For modern, capital-light software or service businesses, relying solely on historical ROIC patterns can be misleading. A software firm might have minimal tangible assets, making its ROIC appear astronomical without truly reflecting a competitive moat. For these companies, the focus must shift to qualitative factors that reveal genuine pricing power or unique scale advantages shared with customers.