agree re: econ's use of calculus (at the undergrad level), since everything gets reduced to a constrained optimisation problem (with potentially slack constrain

agree re: econ's use of calculus (at the undergrad level), since everything gets reduced to a constrained optimisation problem (with potentially slack constraints that require use of Kuhn-Tucker, if one gets fancy). But at the graduate level and beyond, it goes way beyond that. one essentially cannot get into a good phd programme in economics without having done at least up to real analysis as an undergrad. my ugrad profs used to say it is easier to get into grad school in econ with a math ugrad than an econ one. game theory, information economics, social choice theory all make use of lots of pure math. e.g. you have to understand brouwer and kakutani before you understand nash's proofs. discrete math, functional analysis, topology could be useful for 'formal modelling' (which is what the other social sciences call econ-style theorising).