An oscilloscope is invaluable if something dynamic isn't working correctly. All that LCR stuff becomes real when you can see it. SPI bus and i2c bus issues become possible to visualize - though a protocol analyzer is even nicer, a scope will do.
You don’t technically need one, but that’ll make a lot of work much harder and some work impossible.
So no, it's by no means essential but it certainly opens some new doors in the hobby. If you want to build kits, say for guitar pedals (pretty popular electronics-related hobby), you can manage without an oscilloscope.
But for designing experimental circuits from scratch, it's really nice to have one.
And it's just fun to play your guitar with the signal plotted on the warm glowing cathode ray tube.
In terms of needing one: no, if you can borrow one or have access to someone else's lab. Oscilloscopes have relatively low benefit given the cost. They will either be too low frequency to be of value for many modern applications, or too finicky to learn how to use properly for higher frequency applications (not to mention expensive for higher frequency applications).
An AM radio works too, and works above audio frequencies, wirelessly! The loopstick will pick up harmonics off everything in the circuit so you can hear a lot of what's going on. Helps to know what might be going on though.
I had a boss that had retired the following interview question (maybe it was a PhD qual question, I forget): He'd give the candidate a portable TV and an AM radio and have them measure the horizontal scan frequency of the TV. It was probably a chestnut in some circles.
My question is how many people actually use these devices for anything other than trivial purposes these days?
I feel that mine just collects dust and takes up space and does not compare to my multimeter which is probably my most used tool, but maybe I'm missing something.