We assess whether the value of humanity (or global social welfare) has improved in the last decades despite (or because of ) the substantial increase in global population sizes. We use for this purpose a relatively unknown but simple and attractive social evaluation approach based on critical-level generalized utilitarianism (CLGU). CLGU posits that social welfare increases with population size if and only the new lives come with a utility level higher than that of a critical level. Despite its attractiveness, CLGUposes a number of practical difficulties that may explain why the literature has left it largely unexplored. The most important of these difficulties deal with the choice of an individual welfare aggregation function and with the value of the critical level. We address these difficulties by developing new procedures for making partial social orderings over classes of CLGU evaluation functions.These orderings are designed to be robust to choices of individual welfare aggregation functions (within certain classes of such functions) and to ranges of the critical level. The headline result is that we can robustly conclude that world welfare has increased between 1990 and 2005 if we judge that lives with per capita yearly consumption of more than $1,288 necessarily increase the value ofhumanity; the same conclusion applies to Sub-Saharan Africa if and only if we are willing to make that same judgement for lives with any level of per capita yearly consumption above $230. Otherwise, some of the admissible Paretian CLGU functions will judge the last two decades’ increase in global population size to have lowered the value of humanity.