Pretty much everywhere in my field of finance uses some kind of fixed-point representation for good reason. I was wondering about that one day and decided to explore all the things you need to account for when trying to store money in floats.

https://www.moderntreasury.com/journal/floats-dont-work-for-storing-cents

Basically, lots of things go wrong because of the many quirks of IEEE-754, lossy formatting, and various rounding modes. Please use integers, BigDecimal/BigNumber, or even strings instead.