This paper seeks to reconstruct and revitalize the famous Hirschman framewo
rk by providing a comprehensive review of the current use of 'exit, voice a
nd loyalty'. We begin by critically examining Hirchman's original account,
and then look at the way his argument has been extended in different fields
both conceptually and empirically. We suggest that while advances have bee
n made, the results so far are somewhat disappointing given the perceptiven
ess of the original insight. We believe this is because his apparently simp
le schema is more complex than it first appears, and different aspects of e
xit, of voice, and of empirical foundations of loyalty need to be analytica
lly distinguished in order to produce testable empirical hypotheses about t
heir relationships.