Employment brings access to many types of benefits, both financial and psychological. Besides a wage, one can earn a sense of pride or a feeling of connection with her society or fellow employees. How attached one is to the labor market is not an issue that the usual unemployment rate deals with, though. Being unemployed one may not be able to gain access to health care, information about job openings or other non-pecuniary benefits. Knowing a person who is employed may mitigate some of this

more »

... lation but, as recent riots in France and the passing of the National Rural Employment Guarantee in India have shown, simply ignoring the issue of whether someone has access to employment opportunities is not necessarily a desirable option. Taking account of the distribution of employment across households as well as individuals is one way to deal with these issues and improve on the usual unemployment measure. In this paper I develop a new measure of unemployment, that, building on the insights of the literacy and social network literature, is sensitive to the distribution of unemployment across both households and individuals. With this class of measures characterized, I then show some empirical implications and test a main assumption used in creating the measure. JEL Classifications: J64, D63, 047