Patent Linkages and Technology Spillovers in Equity Factor Signals

Summary

This research note explores whether patent data can reveal links between companies that are missed by industry or supply-chain groupings. It uses shared International Patent Classification information to estimate technological similarity, then tests whether returns at technologically related firms help predict later stock returns. The report compares a technology momentum signal with a mean-reversion approach based on return leadership among related firms.

The reported tests cover Chinese equities from 2011 through March 2020. The technology mean-reversion factor performs better than the basic technology momentum factor in the reported information coefficient and portfolio metrics. Incorporating relationships between patent categories further improves reported results, including after removing common style-factor exposures. These are historical, model-based findings; the document warns that the factor may stop working. It provides no detailed implementation, transaction-cost analysis, or evidence that the results generalize beyond the sample.

Key ideas

• Patent classification data can identify technological links between firms outside conventional industry groupings.
• Returns at firms with similar patent profiles may help predict subsequent returns at related firms.
• The report finds a technology mean-reversion signal stronger than its basic technology momentum signal in the tested sample.
• Accounting for connections between patent categories further improves the reported factor results.
• The findings rely on historical model tests and may not persist.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.