This study comprises two meta-analyses conducted to investigate relations between socioeconomic status (SES) and academic achievement, with a focus on macro-level, micro-level, and methodological moderating variables in primary and secondary education. The first meta-analysis is based on 326 empirical studies with 949,699 students from 47 countries and areas, and the second is based on three international large-scale assessments (i.e., PISA, TIMSS, and PIRLS) with 1230 independent samples of 5,095,283 students from 105 countries and areas. We found moderate correlations between SES and academic achievement across the world, rs = .22 ~ .28. Moderation analyses revealed that (a) these relations have strengthened since the 1990s; (b) GDP per capita and economic equality did not affect the relations; (c) higher net enrollment ratio and longer duration of compulsory education did not weaken these relations; (d) the relations stayed stable or even strengthened across grades in concurrent and longitudinal designs. Taken together, our findings suggest that educational expansion that focuses on increasing educational opportunities does not seem to reduce inequalities in academic outcomes between high- and low-SES school children in educational systems on the national level. Quality indicators for educational expansion, however, should be considered in setting educational policy to achieve inclusive, equitable education.