In this paper we discuss how the "encouragement design" used in randomized controlled trials can be extended to a setting with two treatments and one control group. Conditions to interpret the Two-Stage Least Squares (TSLS) estimates causally are stronger than in the case with only one treatment...

For instrumental-variable estimation using binary instruments, we offer simple methods to obtain policy-relevant insights beyond local average treatment effects. We demonstrate the methods by evaluating an element of choice introduced into active labour market policy in Germany by a reform in...