At this time of year, here high up in the northern hemisphere, I find myself longing for the arrival of spring, sometimes poring over the sunset/sunrise times at timeanddate.com . There are a few features of seasonal variation that not everyone is aware of - for example, the earliest sunset of the year occurs two or three weeks before the winter solstice, not at the solstice as many people think.
Now, you’d think the same would apply in summer, but it doesn’t - the latest sunset actually occurs after the summer solstice, in some places weeks afterwards. This leads to the curious fact that the part of the year in which sunset is getting later is several weeks longer than the part when sunset is getting earlier.
You might also imagine that the situation is reversed in the southern hemisphere, but no, it’s the same everywhere. Most of the time, sunset is getting later every day. There’s a thought to cheer you up on a dark winter’s evening.
I imagine it is because the Earth’s orbit isn’t circular and as a result Earth is closer to the sun in the winter (but a couple weeks after the solstice) and furthest during the summer (again, a couple weeks after the solstice); it is also moving faster when closer and slower when further away.
I had wondered this myself, and this nugget of a paper that I dug up on the intertubes is as thorough as you will ever need. Long story short, it’s all about the analemma (which is of course, the result of the tilt of the earths’ axis and of the earths’ elliptical orbit.) Hope this helps.
The answer to your question begins on page 30, but the preliminaries are interesting too.
That paper is from the National University of Singapore, and the case of Singapore shows that I wasn’t quite right to say that this asymmetry of later sunsets vs. earlier sunsets applies everywhere. In places close to the equator, there are actually four cycles of earlier/later sunsets in the year rather than two, and the “getting earlier” cycles are longer.