Let me just quote the ISU on this: "
The sum of the scores for Technical Elements is weighted to count for about 50% of the total score and the five (or three) Program Components will count for the balance of the total score."
This was explicitly stated as one of the principles of the IJS at its inception.
Now, there's the question - 50% of the best skate or 50% of an average skate? I don't think the ISU ever stated this in more detail, but we could certainly go through the different World Championships through the years, and at least get a little bit of an idea.
(And by the way, I am fairly certain that approx. 39.8% of the total score coming from the PCS is the lowest percentage (Ilia's FS at 2024 Worlds) - And 56.2% the highest (achieved by both Weir and Buttle in the FS at 2006 Worlds), though that would require some more digging

)
At the time that the IJS was first put together, cutting-edge jump content for men was three quads (two 3T and one 3S, or vice versa).
And there were 4 spins and 2 leveled sequences.
So look at what the content was ca. 2006 (first IJS Olympics). In the free skate, Plushenko earned 85.25 TES and 82.42 PCS. Buttle 76.80 and 78.50. Lysacek 78.24 and 74.34 and 74.34. Lambiel 76.89 and 76.28.
So yes, the technical content and the program components were more or less balanced -- of the top finishers Lambiel was most balanced between scores.
There was also room for growth in the component scores as judges were reluctant early on to go far into the 8s even for the top skaters and almost never awarded 9s.
Leading up to Sochi in 2014 it seemed there was a push to get judges to be more generous with the PCS. (There was also one fewer spin and a choreo sequence instead of one of the leveled sequences by that time.) So the TES and PCS for the top men's free skates that year were
Hanyu 89.66 90.98
Chan 85.40 92.70
Ten 88.90 82.14
So, quite balanced for the winner, and swings of a handful of points between the two scores for skaters who made technical errors without too much effect on PCS, or who were significantly stronger in general at either elements or components.
But the jump difficulty hadn't changed all that much in the previous decade.
Around 2016, 2017 was when the harder quads started being used more and skaters sometimes started attempting more than 4 quads in a program. By that point, even with the more generous component scoring, there was no way for PCS to come close to the TES of a more-or-less successful quad-filled program. The top 6 men at 2018 Olympics all had free skate TES over 100 -- one of them well over 120 and two others over 110.
Now the available points were no longer comparable between TES (for top jumpers) and PCS. The only way to balance them more similarly to the balance in the earlier years of IJS would either be to increase the factoring for the PCS so the maximum component score would be significantly more than 100, or else to reduce the point opportunities on the TES side. Which was done slightly by shortening the men's program and removing one jump pass.
Once the women started getting in on the quad action, the discrepancy between the top jumpers' TES and PCS was even more drastic because the maximum component score available to women is 80, while the top 2 skaters in the women's free skate at the 2022 Olympics earned over 100 in TES.
Removing one jump pass (in 2026-27) will reduce that gap somewhat, but I still think that if difficult jumps are going to be encouraged the program component factors should be increased accordingly.
And also that more TES points should be available for non-jump skills. And rewarding other kinds of jump difficulty not based on number of rotations in the air.
Not that the ISU has indicated that they intend to go in those directions.