Union membership and age: the inverted u-shape hypothesis under test

In this note we cast some doubt on the claim put forward by David Blanchflower (2007) that the probability of being unionized follows an inverted U-shaped pattern in age with a maximum in the mid- to late 40s. By using a special test for an inverted Ushaped pattern that has not been applied to the age-membership nexus before, and by constructing exact confidence intervals for the maximum value, we demonstrate that at least for West Germany Blanchflower's hypothesis does not hold. Our findings suggest that more definitive evidence is needed before the existence of international unionization-age patterns can be taken for granted.