Occupational gender segregation has generally been assumed to be a structur
e of gender inequality in the labour market; high levels of segregation are
equated with high levels of gender inequality in a society. The paper ques
tions this assumption. It examines, across a range of countries, the relati
ons between United Nations development measures of gender equality and segr
egation levels. Contrary to conventional expectations, correlations are fou
nd to be positive. To explain these results it is argued that segregation,
as measured by conventional segregation indices, is not necessarily indicat
ive of gender inequality which operates to the advantage of men in national
labour markets. The usual segregation measures are argued to be the result
ant of two components: vertical segregation measuring inequality, and horiz
ontal segregation measuring difference without inequality (here vertical an
d horizontal are used in their usual mathematical sense rather than the spe
cial senses sometimes found in segregation literature). It concludes that t
he relationship between segregation and inequality is far more complex than
previously recognized.