The ABO incompatibility is much milder in its (mother/foetus) complications.

It must be, because evolution must have reduced the problem long ago if it ever was more of a problem for reproduction because the prevalence of ABO incompatibility is so much higher than Rh incompatibility. Moreover, in fact, if I remember well, ABO incompatibility protects somewhat against Rh incompatibility, so mitigating the problem and therefore evolutionary speaking promoting ABO incompatibility a little (somewhat like sickle cell trait anaemia (heterozygous) was not much more removed by evolution because sickle cell trait protects against severe malaria.

eritroblastosis is the fetal respond to anemic cause for the hemolisis produced for the Anti D antibodies which destroid fetal cell during intrauterine life.-

Anti A and Anti B inmune antibodies produced neonatal hemolisis instead than fetal hemolisis this is the reason to not found eritroblastosis in that cases but important neonatal hemolisis after 12 to 36 hours after birth

In fact, fetal alloimmunization may be a result from maternal-fetal incompatibility to many blood group systems, not only ABO and RhD (also Kell, Lewis, P1P(K), GLOB, I, among others). Everytime a pregnant women that does not naturally have one of these antigens is exposed to them, there is a risk of development of specific antibodies.

Maternal fetal ABO incompatibility (mother = O, fetus = A, B or AB) may also be a cause of hemolytic disease of the fetus and newborn (HDFN), because maternal Anti-A and Anti-B IgG antibodies cross the placenta. However, most cases of ABO HDFN are generally mild, more prone to occur post-natally, and the reason for that is A and B antigens are not well developed on red blood cells at birth. Conversely, RhD and Kell antigens are usually well developed on the RBC membrane in fetal life, explaining the most frequently seen severe forms of HDFN for these systems.

I support Mr Coutinho explain ethyology of erythroblastosis fetalis.