Working Paper: CEPR ID: DP13533
Authors: Michelle Sovinsky; Eric Helland
Abstract: Every year thousands of firms are engaged in research joint ventures (RJV), where knowledge gained through R&D is shared among members. Many members are rivals leaving open the possibility that firms form RJVs to facilitate product collusion. We exploit variation in RJV formation generated by a policy change that affects the collusive benefits but not the research synergies of a RJV. Estimates from our RJV participation equation indicate participation is impacted by the policy change. The magnitude is significant with an average drop in the probability of joining of 30%. Our results are consistent with RJVs serving a collusive function.
Keywords: No keywords provided
JEL Codes: L24; L44; K21; O32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| leniency policy revision (K40) | RJV participation (R50) |
| leniency policy revision (K40) | likelihood of collusion (L12) |
| likelihood of collusion (L12) | RJV participation (R50) |