Bloom, Nicholas, Mark Schankerman, and John Van Reenen. “IDENTIFYING TECHNOLOGY SPILLOVERS AND PRODUCT MARKET RIVALRY.” Econometrica 81.4 (2013): 1347–1393. Web.

1. R&D Spillovers: The research identifies two main types of spillover effects from Research and Development (R&D). The first is technology or knowledge spillovers, which can enhance the productivity of other firms in similar technological fields.
2. Product Market Rivalry: The second type of spillover is the product market rivalry effect. Unlike technology spillovers, this has a negative impact on a firm’s value due to the competitive threat posed by R&D activities of market rivals.
3. Social vs. Private Returns of R&D: The study estimates that the gross social return on R&D investment is significantly higher (55%) than the private return (21%). This disparity suggests a general under-investment in R&D at the societal level, with the optimal investment being more than double the current observed levels.
4. Methodological Approach: The authors address the challenge of distinguishing genuine R&D spillovers from correlated technological advancements. They use changes in firm-specific tax rules on R&D as instrumental variables, enabling a more accurate estimation of the causal impacts of R&D spillovers.
5. Comprehensive Analysis Framework: The paper develops a framework to analyze both positive technology spillovers and negative business stealing effects from R&D, using various performance indicators (market value, patents, productivity, R&D expenditure).
6. Empirical Findings: Through a 20-year panel data analysis of U.S. firms, the study confirms the existence of both types of spillovers. Despite the negative impact of business stealing, the overall social return on R&D is much higher than the private return.
7. R&D Return Variance by Firm Size: The social returns on R&D are lower for smaller firms, primarily because these firms often operate in more specialized technological niches, creating fewer positive spillovers.
8. Policy Implications: The findings suggest that R&D policies favoring smaller firms might not be the most effective in addressing market failures due to technology spillovers. However, other reasons such as liquidity constraints might still justify support for smaller firms.
9. Further Research Opportunities: The authors suggest exploring industry variations, detailed industry-specific datasets, and mechanisms of knowledge transfer (like trade, supply chains, personnel movements) for a more nuanced understanding of technology and product market spillovers.
10. Methodological Contributions: Despite the need for further research, the methodology presented provides a robust framework for empirically testing the two distinct types of R&D spillovers commonly discussed in growth, productivity, and industrial organization literature.