This study evaluated the extent to which predictions derived from several theories could account for variability in human moral values across US states. We investigated moral values as conceptualized by Moral Foundations Theory, which argues that morality evolved in response to adaptive challenges in at least five domains: Ingroup/loyalty, Authority/respect, Purity/sanctity ("binding" foundations) and Harm/care, Fairness/reciprocity ("individualizing" foundations). We report correlations for measures of cognitive ability, social class, urbanization, pathogen prevalence, life expectancy, and teenage birth rates. Social class and educational attainment had fairly consistent but small effects across moral foundations (social class: positively associated with Ingroup/loyalty, negatively with individualizing foundations and Purity/sanctity; education: positively associated with individualizing foundations, negatively with binding foundations). We conducted multilevel regressions that were stratified for ethnicity. The most consistent state-level predictor of moral values was teenage birth rates (negatively associated with individualizing foundations, positively with binding foundations). This suggests that life-history theory may provide an explanation for individual differences in moral values, although the directions of effects for teenage birth rates diverged from predictions, of life-history theory. We conclude that none of the tested theories provides a good explanation for the observed variability in moral values in the USA. We discuss how a life-history approach might account for the findings, and note the need for improved measurement of pathogen stress to better distinguish its effects from those of life-history variables