Pay-for-delay with Follow-on Products

We study pay-for-delay settlements between a patent holder and a challenger when the patent holder can introduce follow-on products. We show that ignoring follow-on products biases the inferred competitive harm of pay-for-delay settlements (the "Actavis inference"). The reason is that patent invalidation triggers an earlier introduction of follow-on products which changes pay-for-delay negotiation's payoffs relative to the case of no follow-on products. When follow-on products are ignored, we show that an inference based on a reverse payment over-estimates patent strength. If parties cannot use payments (pure-delay settlements), follow-on products may push the parties to settle on an earlier entry date relative to the entry date negotiated in the absence of follow-on products, and litigation may arise in equilibrium