In this paper we develop and test a new set of measures of skill mismatches, based on data on skill levels and skill use in the domains of literacy and numeracy from the PIAAC project. The measures we develop represent the extent of skill use relative to one’s own skill level. We test the measures by examining their relation to a number of labour market outcomes. We subsequently examine how mismatches are distributed across and within a large number of countries, and use our results to reflect on possible causes and consequences of mismatches. We find that, in general, higher skill utilization is always beneficial in terms of productivity and job satisfaction, and that "overutilization" of skills therefore points more towards a fuller use of the available human capital, rather than to a serious skill shortage. We find an asymmetry in returns between literacy and numeracy skills: although numeracy skill level appears to pay higher dividends than literacy skill level, shifts in skill utilization within skill levels have greater consequences for literacy than for numeracy. The distribution of mismatches across and within countries is broadly consistent with the expectation that skills will be used more fully under competitive market conditions with few institutional or organizational barriers. Finally, skill mismatches are only quite weakly related to educational mismatches, reflecting the heterogeneity in skill supply and demand that cross-cuts the dividing lines set by formally defined qualification levels and job titles.