Two important puzzles in epistemology and ethics share a fairly simple metaphysical solution.
Consider these two questions:
(1) How can it be that knowledge is more valuable than true belief? After all, both have the same outcome: the agent who has knowledge of p or truly believes p is equally successful in his dealing with the world.
(2) How can it be that properly motivated action (and action with the right kinds of intentions) is more valuable than action that is not well motivated, or that is only accidentally the right action? After all, these all have the same outcome: the agent brings about an outcome (like giving money to someone in need) that is morally appropriate.
These can look like deep questions that require a lot of epistemology and ethics, respectively, to answer. But that appearance is mistaken. These apparent problems share a common solution, one that does not require complex epistemological or ethical analysis, but rather a fairly simple appeal to metaphysics. The error one is making in taking them to be deep problems lies in a mistaken metaphysics of outcomes. It is, properly speaking, false to suppose that the outcome of true belief and knowledge is the same, and likewise false to suppose that the outcome of properly motivated and non-properly motivated action is the same. We can be led to these mistaken suppositions only if we suppose that an outcome can be characterized without reference to its history. If, on the other hand, outcomes are individuated in a historical fashion, we get a quite different result. Acting successfully because one has a mere true belief, and doing so because one has knowledge, are not identical states of affairs. Bringing about the morally appropriate end result because one has the appropriate motivations, and doing so because of a fluke, are not identical either.
Once we see that these outcomes are different, assigning different values to each is no longer puzzling. It still stands in need of justification, however. In the moral case, the justification is quite simple: it is better to have good motivations than not, so a result that is brought about by good motivations is more valuable than one that is not brought about by good motivations. Once the motivations are seen as part of the outcome, rather than merely part of the outcome's cause, and provided that we admit that having the right motivations is intrinsically valuable, we have a quite plausible justification for valuing properly motivated action more than non-properly motivated action. We also take some of the mystery away from thinking that intentions can make a moral difference, as (for example) in the principle of double effect. (See this post for a more detailed discussion of double effect and consequences.)
A parallel answer can be given in the epistemic case. Knowledge has greater intrinsic value than true belief, so it is more valuable to act according to one's knowledge than according to one's (merely) true beliefs.
The moral: a little bit of metaphysics can go a long way in helping with some puzzles in epistemology and ethics.