First of all, this is nitpicky, and tangential to any normal game play. I know that. If you aren't interested in the question, please to ignore it. However I was crunching a few numbers, found something that bothers me, that I can't answer, and was hoping it might interest someone else here enough to comment.
Take a hypothetical character (probably an NPC) who makes their living from the craft skill. Say that they have a +10 mod, and for the sake of simplicity assume they take 10 on all checks. From their perspective, this allows them to earn an income of 10 GP per week (half the check). So this puts a clear value on their labor.
Now, say they are crafting something with a DC of 20. They will be able to complete exactly one item per week of value 40 GP. According to the rule, raw material costs for that 40 GP item are 13.33 (40 x 1/3). We've just established above that the value of the labor was 10 GP. So where does the other 16.67 GP of value come from? Is that markup costs from the merchant who sells it? Hmm, that actually makes sense. I didn't have an answer when I started typing this, but now I see it. All of a sudden D&D is starting to look like it has a *REAL* economy! Le gasp!!!
Woodcutter (expert 1, 1 rank profession, skill focus, +1 attribute, net mod +5) chops wood, average profession check of 15, and earns 7.5 GP/week. This is profit, there are a few minimal expenses for wear and tear on his tools and such, so he probably actually chops about 9 GP worth of wood each week, to make his 7.5 GP quota.
Carpenter (expert 3, 3 ranks craft, skill focus, MW tools, +2 attribute, net mod +10) buys 13.33 GP worth of wood each week from a couple such carpenters, creates an item (or items) with a market value of 40, but is too busy working to run a shop, so sells it to merchants for 23.33, and there's his 10 GP/week.
Merchant buys for 23.33, and sells for 40. How frequent are sales? Depends on how good of a merchant he is ;) Maybe he only has a mod of +5, and only makes 20 GP worth of sales per week on average (which at that profit margin works out to 9 GP in gross profit, minus expenses on his shop, for the 7.5 GP in net profit that his check indicates). He'll buy more inventory from the carpenter every other week, and the carpenter had better have a second merchant to sell the rest of his stock to.
I already answered my own question, but I'll post this anyway just in case it might interest any other compulsive number-crunchers with a financial bent, who like me are always uncomfortable with the thinly designed and generally-not-so-realistic nature of RPG economics.
Also, it may have some implications toward more realistic costs of hireling labor. Say you want something expensive (3,000 GP "market value", aka list price in the rulebook) made out of wood, say a keelboat. You buy 1,000 GP worth of lumber but don't have time to build it, so you hire a shipwright for 10 GP/week. After 75 weeks (3,000/40) he finishes the project, at a labor cost of 750 GP, and for 1,750 GP you have your finished product. You saved a good chunk of change, and all you had to do was wait over a year. Probably fair. Perhaps if you hired a team of 10 shipwrights you could do it in a couple months at the same costs, and that's still fair. If you actually read everything, including this sentence, kudos to you, and I'm sorry. As a GM you may want to limit the ability to hire teams of too many workers, which is easy enough to do in the name of realism. No, you can't hire 750 shipwrights for 1 GP/day each and expect them to build a keelboat in a single day, to save you almost 50% on the price. Perhaps some sort of inefficiency formula for large teams would be appropriate, but I digress.
A couple flaws in all of this occur to me, from a realism standpoint. First, of course, is the linear scaling of profits from the craft and profession skill. A low-skilled craftsman (say a cobbler), with no mod, probably shouldn't make half as much as a highly skilled (+10 mod) craftsman like a locksmith. The detailed crafting rules do reflect this, as someone with no mod can only hit DC 10 consistently, and thus can craft (10*10) 100 SP, or 10 GP, worth of goods per week. The character with a +10 mod can craft goods worth four times as much, but the "income" portion of the rules suggests he only earns twice as much as his unskilled counterpart. Perhaps these items are more difficult to find buyers for, and so the merchants charge a higher premium, and none of it passes through to the craftsman, but that seems unlikely. If the 0 mod craftsman earns 5 GP a week, producing goods worth 10 GP (and have 3.33 GP of raw material expenses, for a profit margin of only 17%), then the 10 mod craftsman who produces goods worth 40 GP, with 13.33 GP in raw material expenses, should earn something closer to 15 GP, still leaving the merchant a larger profit margin (29%), but also getting a share himself. I'll be playing with formulas for more realistic incomes for higher craft or profession mods, but for now I don't have an obvious answer. Of course none of this applies to someone with a +10 mod who still only makes DC 10 items, but assuming you take advantage of your skills to perform more difficult tasks, it makes sense that there should be some sort of exponential element to the weekly income from a check.
The other flaw is really more of just a common-sense smell test. Can a lone woodcutter really chop down 9 GP worth of wood in a week? Can a shipwright build a rowboat by himself, by hand, in a week and a quarter? Or do any of these results (calculated purely from the game mechanics) produce results completely out of whack with the realistic abilities of medieval craftsmen? I don't know, and it will take research to find out. So in the meantime, I offer you this ridiculously long and useless pontification. Take from it what you will.
