(1) Cost of software is relatively small compared with the cost of employing the people who use it.
(2) Cost of software is relatively small compared with the cost of retraining staff and changing business workflows and processes to cope with different software.
(3) Changing to new software is relatively simple compared with the problems of making all your staff, your contractors and your suppliers change to new software.
(4) Even if changing to new software was relatively simple and cheap, there would be massive risks in doing it. If you are designing nuclear power stations, nuclear submarines, skyscrapers, bridges or whatever, the cost of mistakes can run into the billions, or be fatal for a significant number of people.
(5) Bonus point: making this kind of switch could take a decade and could end up not saving you any money (cf Munich trying to switch to open source). Not many CEOs will attempt it because they are more focused on the next quarter's results. Anyway, in many if not most companies, the benefits -- if any -- would accrue to whoever the-next-CEO-after-the-next-CEO happens to be. And who really cares about that?
It's hard enough to get most people to change their email client or their text editor or whatever when the alternatives are free and the real-world risks are negligible. Getting them to change the software on which their whole business survival is based is another matter.
Moral: changing business software is not as simple as it seems if you only look at the software and don't look at the whole business plus the whole industry infrastructure of related businesses. It's never as simple as people think.