I am an Accounting Ph.D. student and the level of math required in economics puzzles me. Why does economics require so much advanced level math for what is essentially a social science subject? Accounting barely requires basic arithmetic skills, nothing else.
Why does economics require so much advancedlevel math?

I am an Accounting Ph.D. student and the level of math required in economics puzzles me. Why does economics require so much advanced level math for what is essentially a social science subject? Accounting barely requires basic arithmetic skills, nothing else.
What you point out about basic arithmetic skills is exactly why the accounting field is ripe to be taken over by artificial intelligence. I can program a computer to do almost every job done by a CPA. Your classroom will be empty 20 years from now.

Ask yourself this: Does anybody who actually could do something about this  and it’s not obvious that such a person or group even exists  also have the incentive to do so?
Given that advanced math serves as a barrier to entry into the profession, which is good for the incumbents, the answer is probably “no”

We used it to filter out lower IQ morons. It worked well till Levitt discovered that IV+OLS+Amateur Sociology = Guaranteed Top 5. Now we are doomed and are waiting for the apocalypse, so we can restart the whole enterprise
The key is choice, 2cdc. Without it, they will never accept the program.

Why indeed? Poli sci doesn’t. Sociology doesn’t. And as the powers that be have stated, they wish to move econ in the direction of sociology, since sociology doesn’t have a gender problem. So the next logical step is...
HRM polisci programs only admit people with at least a calculus sequence and probabilitybased stats. I know that's not real math, but this isn't comparable with sociology, which takes anyone with stats 101.

Math is a useful, precise, and compact language for expressing ideas in a consistent manner. Read some of the work of the marginalists if you want to see how much easier it is to understand basic things like supply and demand with math. That illustrates the value of a compact language. To see the value of precision, read any of Williamson’s stuff and try to answer any basic questions about it without one of his interpreters in the room.
That all explains why math is a useful language, but why the complicated math? It turns out that a lot of ideas can be conveyed cleanly with high school math. But as you interrogate these ideas more and more, you’ll see that the only way to do so is to look at their generalizations. And those, for better or worse, involve more difficult mathematics. For example, you might ask what happens if utility functions are not differentiable, say, because there is some discreteness in the underlying choices (as there often is). That requires more complicated tools. Or what if there are shocks that are neither discrete nor continuous (say because a privately informed competitor’s entry choice is discrete, and their pricing decisions are continuous)? Then you need a little bit of measure theory. Or what if you know your model isn’t exactly right, but it’s pretty close, and you’re wondering about whether it’s predictions will nevertheless be pretty close? Then you need a little bit of analysis.
Some people like more complicated math because they think that if they can use it, it makes them look smart. Ignore those people. They’re just playing a game. More complicated math is useful because it’s often the easiest way to express internally consistent ideas in a way that captures their depth.

Math is a useful, precise, and compact language for expressing ideas in a consistent manner. Read some of the work of the marginalists if you want to see how much easier it is to understand basic things like supply and demand with math. That illustrates the value of a compact language. To see the value of precision, read any of Williamson’s stuff and try to answer any basic questions about it without one of his interpreters in the room.
That all explains why math is a useful language, but why the complicated math? It turns out that a lot of ideas can be conveyed cleanly with high school math. But as you interrogate these ideas more and more, you’ll see that the only way to do so is to look at their generalizations. And those, for better or worse, involve more difficult mathematics. For example, you might ask what happens if utility functions are not differentiable, say, because there is some discreteness in the underlying choices (as there often is). That requires more complicated tools. Or what if there are shocks that are neither discrete nor continuous (say because a privately informed competitor’s entry choice is discrete, and their pricing decisions are continuous)? Then you need a little bit of measure theory. Or what if you know your model isn’t exactly right, but it’s pretty close, and you’re wondering about whether it’s predictions will nevertheless be pretty close? Then you need a little bit of analysis.
Some people like more complicated math because they think that if they can use it, it makes them look smart. Ignore those people. They’re just playing a game. More complicated math is useful because it’s often the easiest way to express internally consistent ideas in a way that captures their depth.tl; downvoted

PAtological verbosity
Math is a useful, precise, and compact language for expressing ideas in a consistent manner. Read some of the work of the marginalists if you want to see how much easier it is to understand basic things like supply and demand with math. That illustrates the value of a compact language. To see the value of precision, read any of Williamson’s stuff and try to answer any basic questions about it without one of his interpreters in the room.
That all explains why math is a useful language, but why the complicated math? It turns out that a lot of ideas can be conveyed cleanly with high school math. But as you interrogate these ideas more and more, you’ll see that the only way to do so is to look at their generalizations. And those, for better or worse, involve more difficult mathematics. For example, you might ask what happens if utility functions are not differentiable, say, because there is some discreteness in the underlying choices (as there often is). That requires more complicated tools. Or what if there are shocks that are neither discrete nor continuous (say because a privately informed competitor’s entry choice is discrete, and their pricing decisions are continuous)? Then you need a little bit of measure theory. Or what if you know your model isn’t exactly right, but it’s pretty close, and you’re wondering about whether it’s predictions will nevertheless be pretty close? Then you need a little bit of analysis.
Some people like more complicated math because they think that if they can use it, it makes them look smart. Ignore those people. They’re just playing a game. More complicated math is useful because it’s often the easiest way to express internally consistent ideas in a way that captures their depth.