In the social policy debate, fundamentally different ideas prevail about the interlinkages between such key variables as employment, low pay, social transfers and poverty. This paper presents basic empirical evidence on the validity of these ideas and the policy prescriptions that follow from them, mainly drawing on cross-country comparative analysis. We show that clear and striking cross-country correlations prevail, but not, as is often so readily suggested, between low pay (wage compression) and employment performance, or between employment performance and poverty. Instead we find a strong and positive cross-country correlation between the incidence of low pay and the incidence of relative poverty, and we also find a strong but negative cross-country correlation between the level of social spending and the incidence of poverty. In addition, the incidence of low wage employment and social expenditure are also strongly and (negatively) related. We examine these correlations in more depth, particularly the link between the level of social spending and poverty. Since there is such a clear and strong negative link between the level of social expenditure and the level of poverty, it is tempting to think that more social spending offers an easy route to less poverty. However, a simple simulation exercise using Luxemburg Income Study data from the mid 90's suggests that putting more money in social transfer systems as they currently exist in the EU would not have positive outcomes on poverty rates in all countries. The final section of the papers sets out an agenda for further research.