I think there are several misconceptions here, especially surrounding
most 2000s-era anime and movies were done digitally which means direct-to-disc
but I can't speak with any authority about the production workflow of studios.
I can speak with at least a little authority when it comes to video/encoding, so I'll try to stick to that.
Up to the late 2000s, in regions that followed NTSC (as opposed to PAL or SECAM), pretty much all consumer TVs only had analog inputs (rf, composite, s-video, component video). They only had the hardware to display 30 frames per second (...almost, I'm sure I'll get into that later) and expected the signal to be interlaced, displaying 60 fields per second.
Since that's what the hardware supported, that's what got put onto DVDs. Japan used the NTSC broadcasting standard.
While I can't say for certain that no studios produced video at 30fps, I'm pretty comfortable saying that most produced (and still produce) video at 24fps. So there is likely no "native" 30fps. It's all just 24fps with a bit of trickery to turn it into 30fps.
The Long Version
Pretty much anything involving video is going to have an incredible amount of detail and math. I'll try to restrain myself.
I'm gonna go into some amount of detail about the history of video/film as background, because knowing that is going to be helpful to understand why "native 30fps" isn't really a thing for anime.
TVs have evolved over time
We started with black&white broadcasts at 60 fields per second, matching the frequency of AC power. The reason why it's fields and not frames appears to be due to slow phosphors in the first TV sets. Hitting every other scanline eliminated strobing effects.
Each frame would have its "even" numbered lines (the "even" field) scanned in the first pass1, and then the CRT beam would make its way back to the top of the screen so that it could scan the odd numbered lines (the "odd" field) to display the entire frame.
Color TV came along, and for a bunch of math reasons I won't go into (partly because I don't understand them), the framerate had to change so that the old black&white TVs could still display the video (without noise/artifacts appearing). This is why we have framerates of 29.97 (or rather,
30,000/1,001) for "video" and 23.976 (
24,000/1,001) for "film". This video from Matt Parker of Stand-up Maths goes into a bit more detail.
Because they're close enough to whole numbers, I'll just be referring to it as 30fps/24fps from here out.
The point here is that we're still stuck with the consequences of practical decisions that were made 70+ years ago.
DVDs are a standard, meant for playback on TVs
The DVD format is standardized (a process that takes megacorps years to hash out amongst themselves and results in a lot of long, extremely technical books). You can't just throw whatever you want onto a DVD, at least not if you expect to be able to play it in a standalone DVD player.
They should all be 480i30 (720px wide x 480px tall, interlaced, 29.97 frames per second)2 and encoded with MPEG2 (Main Profile & Main Level) and packaged as a Program Stream because (as far as I know) that is what the DVD Forum decided on.
While DVD is itself digital, it had to output an analog signal so that it could play on the TVs that people had. It's hard enough to get people to adopt new tech. It'd be near impossible to convince them to get a new TV too (at least in numbers that would make DVD financially viable).
Film and Video
Film, like movies out of Hollywood, that kinda film, settled on 24fps.
Video, like TV broadcasts, settled on 30fps
Watching movies at home wasn't a thing until the 1970s, but when it became a thing, that framerate difference was going to need to be handled somehow. TVs of the time couldn't do 24 fps (heck, I think that having 24fps processing in TVs didn't really become widespread until the 2010s). So how are we gonna convert 24fps to 30fps?
The answer is Telecine (I believe in industry it rhymes with "whinny", like the sound a horse makes, or Whinnie the Pooh. No source on that.), a.k.a. 3:2 Pulldown.
You take 4 frames, split them into 8 fields, and then strategically duplicate some of the fields so that it blends together relatively smoothly. Honestly, the Wikipedia article on Telecine/3:2 pulldown probably does a better job showing it than I could hope to do. You end up with 5 frames for each of the 4 original frames, and 24 * 5 / 4 = 30. Problem solved!
Getting back to DVDs for a bit, you could do the 3:2 pulldown before encoding the video with MPEG2, but I believe that MPEG2 has a "flag" where you can encode at 24fps and have the playback hardware handle the field weaving.
This is why you can't just take the 30fps video you get when you rip a DVD and simply convert it to 24fps. You're going to get interlacing/combing artifacts if you try that. You need to de-interlace (or rather IVTC, Inverse Telecine) to reconstruct the original, progressive frames3.
As far as I know, studios produce at 24fps. That was the case then, and it's still the case now. If I were to speculate, I'd say this is motivated by cost (fewer frames = less money to make) and inertia (it used to be done at 24fps, the equipment they have works at 24fps, and it would be $$$ to replace that)
There may be some series that have title/credit sequences at 30fps and everything else at 24. Doing any amount of processing on those is a pain.
Japan did use NTSC as well (NA = region 1 coded on DVDs, JP = region 2. Geographic region-locking has plagued us well before Netflix and the fragmentation of streaming sites), so 24fps telecined to 30fps is what the DVDs coming out of Japan were.
Because of that, regions using PAL could (or were more likely to) get the short end of the stick. Releases in PAL regions (DVD regions 3, 4, 5) can just blend frames together4 to get to the 25 frames per second (50 fields per second)5 needed to display video on their TVs. They could also simply speed up the video (and pitch-adjust the audio). The method that gets used would depend on the producer/distributor and could change as time moved forward.
Beyond that, there isn't that much merit in reducing the framerate for DVDs. Single-layer DVDs can hold about 4.7 GB of data. Video takes up a lot of space. 24fps, 720x480, 32 bits per pixel6 means a single frame of video, uncompressed, is 1.38 MB. A single second is 33.17 MB. MPEG2 doesn't employ a ton of tricks to reduce the filesize like h264/h265/av1/etc... do. A typical episode off a DVD will be around 1 GB. Everything else being equal (it's not, but let's pretend it is), you'd be able to squeeze 5 episodes onto a single DVD instead of 4.
There's some way that distributors have been able to get 7 episodes on a single DVD (see the Hyouka blu-ray + dvd part 1 combo for an example). I assume that's because they're using a dual-layer disk instead of a single-layer one (which means you either wouldn't be able to see all the episodes on an old enough player/drive, or they may not be able to read it at all).
Computers/TVs have become a lot more flexible after the rise of HDMI, and processing power has also increased. As a result, Blu-Ray is also more flexible (capable of storing interlaced or progressive video, at a number of resolutions, at a number of framerates). If you find yourself with a blu-ray for an anime that's at 30fps rather than 24fps, that's probably a sign that it's... not a great release of that particular series).
Thank you for reading my TED talk.
1: even/odd fields can also be called top/bottom. Scanning can be done TFF (top field first) or BFF (bottom field first), but that detail doesn't really matter much for this discussion.
2: yes, I know that PAR, SAR, and DAR are a thing. As well as "active lines". Not gonna touch either of those here.
3: from my hobby editing days, I also recall a deinterlace + decimate process where the deinterlacing would reconstruct 5 frames (one being a duplicate), and you needed the additional "decimate" step to remove the one duplicated frame in the set. I can't quite remember the exact details. There were several different chains of AVISynth filters you could use to get from the .vob files you got from a DVD to something you could work with in an editing program.
4: see this section of the "Analyzing your DVD footage" guide on animemusicvideos for some more detail which is specific to anime (as opposed to, say, this link to doom9 which is about video processing in general)
5: yes, actually 25fps and 50 fields per second for PAL. They saw the mess with NTSC and specifically crafted their standard so that the framerate would be an integer
6: 32 bits/pixel is for RGB. MPEG2 uses the YUV colorspace with 4:2:0 sampling. That's one of the "tricks" they used to reduce the size of files. They literally discard some of the color data because our puny human eyes/brains care more about the difference in bright/dark and we'll be none the wiser... for the most part.