The main reason is that occasionally some disciplines decide to change their style for their own benefit. The one I am most familiar with, mathematics, used to footnote citations until about 1945. Typesetters hate footnotes and charge extra for it. I don’t know if that was the reason but about the end of WWII, a few papers began to use a style of using numbered references that were cited by a number in brackets: . This was much less informative, but by 1950 it had taken over completely. Now a few people are experimenting with citations of the form [Author, date], which is very informative, even it does take up a bit more space. This is standard in linguistics and presumably some other disciplines. So the basic reason is that each discipline sets its own style and there is a lot of momentum against change.
I just read a book that uses endnotes for citations and the endnotes refer to a bibliography. So if you want to track down a citation, you first have to find the endnote (their numbers start over in each chapter, so you first have to figure out what chapter you are reading and then find the endnotes for that chapter) and then, after you finally have the name, you have to go to the bibliography to find the actual publication. I find this system awful; books should be written for the convenience of the–presumably many–readers, not the one writer.