The identification of average causal effects of a treatment in observational studies is typically based either on the unconfoundedness assumption or on the availability of an instrument. When available, instruments may also be used to test for the unconfoundedness assumption (exogeneity of the...

This paper examines changes in the distribution of wages using bounds to allow for the impact of non-random selection into work. We show that bounds constructed without any economic or statistical assumptions can be informative. However, since employment rates in the UK are often low they are...