Expanded international data from the PIAAC survey of adult skills allow us to replicate the IALS-based analysis by Coulombe, Tremblay and Marchand (2004) as well as Coulombe and Tremblay (2006) based on more recent and more comprehensive data on the literacy skills of the adult population. Results from panel estimations over the period 1970-2010 suggest that literacy skills have become an even more important determinant of economic growth than was suggested by the IALS analysis covering the period 1960-1995. Our estimates imply long-run elasticities of GDP per capita with respect to literacy of about 3. This means that in the long run a one-percent increase in literacy translates into a three-percent increase in GDP per capita. Short-run elasticities are also substantial. The association between labor productivity and literacy is equally strong. This suggests that the effect of literacy on living standards goes beyond its effect on unemployment and participation rates A closer inspection of the data additionally reveals some important heterogeneities: Investment in the human capital of women appears to have a much stronger effect on subsequent growth than investment in the human capital of men. Our results also suggest that underinvestment in human capital hampers growth by more than developing highly talented individuals stimulates it. Specifically, the proportion of adults with low levels of literacy skill -- Levels 1 and 2 -- appears to have a much larger impact on growth rates than the proportion of adults with Level 4 and 5 literacy proficiency. Thus, policies that serve to reduce the proportion of low skilled adults would likely yield higher returns than those that serve to increase the proportion of high skilled adults.