Why add baking powder when creaming sugar and butter?

I can't think of a reason to add the baking powder before creaming, but I can think of a few reasons not to:

Pretty much all baking powder these days is "double-acting". It provides leavening two different ways, under two different conditions: first, when it comes in contact with moisture, it produces carbon dioxide, and second, when it comes in contact with heat (in the oven). Adding the baking powder earlier than usual in a cookie recipe would trigger that first leavening earlier than normal, as soon as the baking powder comes in contact with the water in the butter. It's normally added at the end of the process so that as much of the CO2 as possible from the first leavening is still in the dough when it goes in the oven.

Perhaps more importantly, adding the baking powder to a creamed cookie dough very close to the end of mixing, along with the flour, keeps the agitation of the beater from beating too many of the carbon dioxide bubbles out of the dough. That's normally the first time the BP encounters moisture and starts that first leavening, and ideally not much mixing takes place after that step. Adding it before the creaming step begins would mean that any CO2 bubbles introduced by the first action of the baking powder aren't going to stand much chance of survival through the rest of the process, as the butter and sugar are creamed, eggs beaten in, etc.

Also, baking powder doesn't last forever, its leavening power decreases with age. Yours may just be old, or at least old enough that those two other factors were enough to affect the end product.

I was thinking about this for a moment. Looked at few recipes and found out that there are few where the baking powder and sugar are mixed before adding the butter.

All of them used "I'm damn sure this is not a butter". So margarine, butter and hemp mixture, cocoa butter. Also one of them said about beating the mix to foam. So that, as mentioned by Dan C, may be the reason to add something that will bubble.

I recently used a recipe from King Arthur Flour for cinnamon streusel coffee cake with the same instruction. I've never seen any other recipe call for it, but I did it. The result was a cake with only a few fully-raised places. Ugh.

Also, as I understand the chemistry, it is not moisture per se that causes the initial rise from the baking soda component in baking powder, it is acidity. Butter is, usually, slightly acidic, so as noted above, would begin the activation of the BP - and the creaming process, plus the beating in of the eggs, would drive out nearly all the potential leavening.

(A further note on the eggs: they will also cause activation if they are added unbeaten, as yolks are slightly acidic and activating what they hit; if beaten, no net effect since the whites are slightly basic and the net basically neutral except for very old eggs)