Except now it's stopped at $320. That's an average of $93.33 per logo (if my math is right). They would make more money by simply selling the logos for $100 each, and they would probably sell way more logos.
1 - Start pricing extremely low, and word travels fast
2 - Temporary pricing causes buyers to act quickly, with less consideration
3 - Buyers who wavered in their decision making panic and pay
more than they originally would have in order to not miss out.
4 - Designer knows what is left is both over priced and no one really wants.
With static pricing, #1 doesn't happen, #2 doesn't happen, #3 doesn't happen, #4 does, albeit much more slowly.
#1 is a core piece of both Google & Facebook's sales psychology. You can see them move the sliders from totally free, to really good deal, to your fucked without paying us lots of money as they launch ad channels.
If he sells all 50 that he has, he'd net an average of $1020 per logo. (First one was $40, price increases by $40 per logo, last logo is $2,000, total net is $51,000)
That's a highly unlikely "if". I doubt they'll sell any logos for much more than $400, which averages "only" $146.66 per logo. That's pretty good one-off revenue if the logos were already finished, especially considering they can probably sell the remainders on the side for some extra cash, but it's probably not a sustainable business model.
That does seem to be the case. It actually could be an interesting sales move. If the price would rise linearly, the saturation point in which the logos become "too expensive" would occur faster in sales. Gently lowering the additional value (from 40 to 20 and i assume from 20 to 10) would let the saturation point occur later, thus having more sales in total.