The only thing I can think of is that as the reinforcement ratio increases beyond a limit the level of ductility/curvature you might ultimately get goes down. Our local code has a lower limit on the maximum tensile reinforcement ratio for potential plastic hinge regions where more local ductility demand is required. I think from memory its 0.025Ag, but where there is essentially no local ductility demands in areas of limited ductility (read as less curvature than 3-5 times the initial yield curvature) a limit of 0.08Ag can apply for the tensile reinforcement. As you can appreciate its a very large difference in reinforcement content.
Additionally for a higher ductility the beam section will behave more dependable if there is higher ratios of compression reinforcement compared to the level of tensile reinforcement (as this essentially limits the compression block depth, which improves the curvature that can be sustained), again in our local code there are minimum ratios on the level of compression reinforcement for potential plastic regions. This is stated as some ratio of tension reinforcement ratio. From memory its a lower limit of 0.38As or 0.5As depending on the level of local ductility demand (As = tension reinforcement area). There is no upper limit on the level of compression reinforcement, but for reversing hinges its inferred as being the inverse of these ratios, as when the beam is bending the opposite way under reversing seismic loads the provided reinforcement also has to satisfy these ratios (tensile reinforcement under reversal of load becomes compression reinforcement, etc).
The way our code works in NZ is to assess the level of curvature occurring at potential plastic hinge regions from the final deformed shape. This level of total curvature then categorises the hinging regions as NDPR (Nominally ductile plastic regions), LDPR (Limited ductile plastic regions) & DPR (Ductile plastic regions), and each region has different detailing requirements regarding maximum/minimum reinforcement requirements for both longitudinal and shear reinforcement.
You still need to satisfy balanced strain criteria for the cases where compression reinforcement is used, I say this because I have seen some people say as soon as you add any compression reinforcement suddenly you don't have to worry about balanced strain, this is not correct.
