That's because without DST as the day lengthens the extra daylight is split evenly between morning and evening.
But WHY does it split evenly? It's because we historically use noon as our reference. We think of noon on day N+1 as being exactly one day after noon on day N. Noon is a sensible reference, fairly easy to determine with ancient technology.
Suppose that instead we had historically used sunrise as our reference. By definition, then, sunrise would be 8 AM every day. As the days lengthened in the summer, all of the extra daylight would then have shown up in the evening automatically.
It's interesting to ponder whether or not such a reference could have worked. I think that in the days before clocks, when the sundial was the most sophisticated timepiece, this could have worked quite well. It would also fit in well with work schedules, meaning people who needed to wait for light to work would be going to work at about the same time every day. It would also fit in with animal sleep schedules.
With clocks, though, a noon reference is easier. A sunrise reference requires a much more sophisticated clock, or it requires frequent manual adjustment.
Nowadays, though, clocks easily have the sophistication to handle that. We probably could actually reasonably handle using sunrise as our reference, and have our clocks adjust daily to keep that so.
It would mean that a given clock time on day N+1 is no longer exactly 24 hours later than that same clock time on day N, but I don't think that would matter for ordinary life. I'd be willing to put up with that if it means I get the benefits of DST (more evening light in summer) without the two big discontinuities a year.
Edit: to be clear, I am assuming that if we did a sunrise-based system instead of a noon-based system we would still do time zones. 8 AM would be simultaneous everywhere in the zone.
[1] In this comment "day" refers to sunrise to sunset.