Once thousands upon thousands of missiles are being mass manufactured, the cost drops drastically. Down to the point that the only limiting factor is the raw materials cost and maintenance cost for the manufacturing line. From what we can tell, all of these missiles are made on a robotic assembly manufacturing line. No human hands on them. The cost is all tied up in R&D of the missile itself, the set up time of the assembly line, and lastly, the minor part, the raw materials.
Same reason JDAM kits cost so little. Initially it was thought it was going to be really expensive, then the secondary test results came back(after proof of concept testing). An even 30 for 30 hits. Well was 28 for 30, but was found that the missed container targets were literally MOVED by the shockwaves from the other 28 dropped 2000lb bombs thus creating a "miss". The Airforce immediately told Boeing(bought McD's), who really was nothing but a name plate holder while Honeywell did all of the real work, to mass produce them ASAP. Thousands of kits were manufactured. The cost plummeted. Why JSOW, etc are so expensive. All that R&D sunk cost, but the airforce and navy are not really buying.
On the other hand ExCalibur is expensive, effective, but not very useable. Why? Range limited to 45km or 30 miles or so. Initially it was tied to the hip of Paladin going with +200mm artillery instead of the 155mm standard. Anyways, the larger artillery would have given a radius of over 60 miles increasing its area suppression, denial, effectiveness by 400%. So, very few were ordered as the final product, the 155 round, has so little explosive in it, that effective suppression is compromised. (personally, I would ditch 155mm and go with a combined +200mm, and 100mm or 120mm for commonality with tank ammunition).
For instance, mistletoe will be WAY more expensive than missiles. Even though it is a simple modification to an RD architecture. Hmm, wonder how many of these buggers the RMN bought?
Could destroy my thesis right there huh?