Markit Bows Dataset of Companies' Patent Value

Markit has unveiled Markit Intellectual Property Model, which assesses the value of publicly-traded companies' patent portfolios, based on information from patent analytics and research provider The Patent Board, to enable users to use the inherent worth of companies' intellectual property as an additional input to support investment decisions.

The Markit IP Model comprises nine factors, including metrics on patent production efficiency -- which takes into account the number of patents a company generates compared to industry peers, the size of the company or the amount of its investment in research and development -- citations and scientific relevance, as well as the changes in those metrics over time. Mike Stritch, director of Markit's Data Analytics and Research group -- formerly Chicago-based transaction cost analysis and over-the-counter factor-based research provider Quantitative Services Group, which the vendor acquired last year (IMD, Nov. 5, 2011) -- says this can help traders, analysts and portfolio managers predict future returns.

"Citation trends-meaning whose patents are getting cited the most-is a measure of quality and can lead to future economic profit for that company," Stritch says. "Some industries favor more innovative, newer technology, and other industries favor hard science, which we try to capture by looking at the number of scientific references that are made."

Markit developed the new model in partnership with The Patent Board, which collects and analyzes patent-level details on over 4.5 million filings from the US Patent Office, and generates company-level patent data by linking patents to products, revenues and scientific papers, creating a model that can be leveraged to evaluate publicly-traded companies, Stritch says.

The vendor provides transparency into the underlying data and factor models -- which users can easily compare to any of Markit's other equity factors using its factor analytics platform to determine correlations and to see how these factors measure up against each other -- and provides 10 years of history for the factor suite, enabling users to back-test their strategies. "In certain industries, having the highest score alone adds value over time.... Or maybe you look at measures to find residual trading opportunities. Maybe you have a company that's trading at a premium but has a relatively low IP portfolio, which might be a sign that the valuation is off. On the flip side, a good valuation and a correspondingly attractive IP portfolio could make the company a potential buy candidate," Stritch says.

Besides quantitative traders, analysts and portfolio managers that see these factors as potential new inputs to their models, fundamental data users can also leverage these factors as a way to systematically monitor companies' IP position over time, Stritch says. "If you're a tech analyst, you need to be aware of patents. If a company issues 300 patents in the last 12 months, it's hard to get an overall picture, so this provides a nice set of inputs to consider against valuation and future cash flow potential. We're not offering individual patent-level details, but portfolio-level, which would otherwise escape the attention of fundamental traders," he adds.