You're missing the point of marginal utility. Because the price of water is so low, the marginal use of water is not for purposes vital to survival. People water their lawns, fill swimming pools, wash cars, etc. In other words, they use water for things whose utility is equal to its current price, because there's plenty of water left over at that price after all the vital-to-survival needs are met.
If the price of water went up, people would stop using it for purposes with lower utility, in order to be sure to have enough for purposes with higher utility (like drinking and cooking). So the marginal utility of water would still be equal to its price.
See David Friedman's text, Price Theory; in particular Chapter 4:
An introductory text - thank you, but I've taken microeconomics in college. After carefully considering both points of view, I tend to rather agree with Smith, for the following simple reason. Suppose the more frivolous uses of water were somehow restricted. (There are cities that do pass such regulations sometimes.) Its marginal utility would increase, but its price would most likely drop, due to lower demand. Alternatively, consider that blood plasma and many life-saving medicines that have no other use are still cheaper than diamonds.
In any case, let's go back to the original issue. And, for the sake of argument, let's agree with the marginalists. Aren't all the marginal values and costs for a digital creation, which can be easily replicated millions of times, practically zero?
> Suppose the more frivolous uses of water were somehow restricted.
Then you would no longer have a free market; people would be prohibited from engaging in transactions that they would have chosen to engage in without the prohibition. In that case, yes, price is no longer equal to marginal utility.
> Alternatively, consider that blood plasma and many life-saving medicines that have no other use are still cheaper than diamonds.
A better example, yes. I would tend to say the "market" in these things is not exactly free either. Somebody should ask Friedman about that one. :)
> And, for the sake of argument, let's agree with the marginalists. Aren't all the marginal values and costs for a digital creation, which can be easily replicated millions of times, practically zero?
And that's part of the point ... These aren't typically free exchanges. Part of the complaints brought on my customers is monopolistic (oligopolistic) practices, including forced bundling, false shortages, region restrictions, etc., coupled with government limitation (e.g., copyright laws).
Additionally, (and this is a point I argue with those proposing wholly market-based solutions to public education) a theorized free exchange with unshackled supply/demand curves assumes perfect information, which we don't have. (In this case, partly due to the long tail of attention.)
I'm pretty sure the basic assumptions of the economics y'all are discussing are not met because they require concave production functions. According to micro theory (I'm trying to recall Mas-Colell here so maybe way off), a linear production function as we have with internet distribution results in corner solutions of 0 or infinite production, and either way can't be priced (taking record production to be a sunk cost, which it is by common definition).
Point being, any argument that relies on a micro framework is bunk because we've reached a level of technology where convex production sets are no longer an obvious assumption.
If the price of water went up, people would stop using it for purposes with lower utility, in order to be sure to have enough for purposes with higher utility (like drinking and cooking). So the marginal utility of water would still be equal to its price.
See David Friedman's text, Price Theory; in particular Chapter 4:
http://www.daviddfriedman.com/Academic/Price_Theory/PThy_Cha...
He discusses this exact case (diamonds vs. water).