They deprecated 32 bit support almost a decade ago. Software has a lifespan and always has, to run obsolete code you need to jump through hoops to get an obsolete target running. This isn't new.
Why is 8 years self-evidently "long enough"? Banks have code from the 1960s still running, and plenty of people have industrial hardware much older than that. Many of my household appliances are older.
It's only "obsolete" code because Apple has decided it is, it didn't rot.
The only reason "old code" still works in banks is because untold man hours are spent wrapping around it, working around it, and testing it.
A long time ago, some exec saw that some old thing needed updating because the world has changed. There are new regulations, new customers types, new products, new strategies...
A bunch of people get hired. A bright future of next thing is promised. It gets everyone excited and devs start cutting little chunks away off the old thing, making a new thing. Now there are 2 things to maintain, but that's ok, some day new thing will become next thing.
But the exec leaves (probably to another bank) before the next thing became a thing. Another exec comes in, declares the new thing is crap because it doesn't quite get the job done. A next next thing is planned. The previous new thing is now the old new thing. It will be removed, someday. Just not today, because it's _a_ thing now. old new thing will be sunset when next next thing arrives.
Many many years and several execs later, there are many things maintained by a small army of devs and testers. These things work in really strange, archaic, or even stupid ways. Nobody knows how (or worse, why) the entire system works. The most knowledgeable people only know how it _behaves_.
But it's there, still working. And your money is in that thing of things. Many peoples' money are in it.
That's not necessarily a good thing.
That said, I don't agree with such an argument. It seems like seeking out technical excuses for crappy user outcomes. In another thread on this topic I said: in any job there are parts that are fun and others that are not so fun but necessary. They should try to keep customer apps working.
The Mini was the last Mac to make the switch to 64-bit, in "Mid 2007". So a little over 12 years.
For comparison, the Macintosh was on M68K from 1984 to 1994 (10 years), and on PowerPC from 1994 to 2006 (12 years). IOW, the Mac has been on 64-bit x86 for as long as they've been on any CPU architecture ever.
Another fun fact: there was less than 10 years between the last Apple IIe they sold and the first Power Mac G5.
The 32 bit support is actually a bit of a red herring here; the bigger issue for some is Apple dropping support for Carbon, a 32 bit only API made to ease porting of classic MacOS apps from the 80s and 90s to OS X, and itself deprecated in 2012.
I guess Apple could have created a 64 bit version, but it also seems reasonable to expect software developers to have rewritten effected software in that timespan. Carbon was originally a compromise for companies not wanting to learn an entire new OS (NeXTSTEP) and API/libraries when updating their apps for OS X -- the fact some apps still need it today is kind of mind boggling.
Then, I suppose the Win32 API has been stable-ish for far longer.
There is lots of great modern art. but none of it makes older art "obsolete".
If a painting sits in a moldy basement for a decade, don't be surprised when someone finally go to check on it at what happened.
In other words it's not about the overhead, it's about the reliance on x86.
It seems kind of silly to block x86 modes on x64 chips when the backwards compatibility needs of Windows pretty much insures "eternal" x86 modes. It's going to be a while before x86 modes disappear on Intel or AMD chips, and there's evidence to support that like the Itanium (IA-64) failure, and the fact that Windows on ARM made it a commitment to emulate x86 (but not x64) on ARM. x86 is likely going to be a lasting "lingua franca" ISA. Right now I'd almost put more money on x64 being retired before x86, and if anything from that perspective Apple is more tightly coupled to an ISA than before this choice.
That really depends on what their long-term goal is. If the speculation & rumors about an eventual ARM transition have any merit at all, then sunsetting 32-bit support now can help them smooth that transition, since they'll have a lot less work to do to implement whatever Rosetta-esque dynamic binary translator they would need.
Which it seems unlikely that Apple could do in secrecy what Microsoft + Qualcomm could not do (very publicly), but I guess all things are possible.