For what it's worth I ran the following test several months ago. The purpose was to test the lightfastness of the most vulnerable colors, the brilliant reds and oranges of multiple pastel brands. I did not record the identifying names/numbers of the sticks, just the brands. I included Holbein watercolor as well since I use that for underpainting.
I purposefully did not test NuPastel because I know there are lightfast issues with reds in that brand. (in retrospect, I wish I had included them for comparison)
First, I made solid strokes of color on a piece of mounted Wallis paper. Then I covered the bottom half of the marks with a piece of matte board, taping the edges with black tape (the black tape did not touch the pastel). Here is what it looked like:
I placed this in a south facing window that received partial sun.
After three months, I took it down and removed the cover:
As you can see, there is virtually no change. This surprised me but made me feel more secure about using these pastel brands. The only one I noticed was the Great American yellow-orange and the Holbien watercolor Carmine, both of which showed only a very slight change, not really enough to concern me.
In addition, I ran the same test using the UArt sanded paper (dull yellowish color) that is currently available. Please note, this is NOT the same UArt sanded paper (light beige color) that will be introduced soon.
As you can see, it is subject to fading. Don't get me wrong, I LOVE UArt paper, but if I am going to leave transparent areas or watercolor underpainting showing, it is helpful to know that things may change over time.
(Again, this is NOT the new UArt paper that will be shown at IAPS which is more beige in color. I am currently running another test on that one, but it will be a while before I know the results.)