I know that if someone is really interested in copying my information will eventually find a way to do it, but I find that it works great as a deterrent for most scenarios.
An attacker is going to take the path of least resistance. Are they going to spend several hours trying to edit out your watermark and end up with a result which might still be rejected by whichever service or bank they are trying to fool, or are they just going to move on and use someone else's KYC data instead? When KYC data is widely available to be bought on black markets for ten or twenty bucks for hundreds of users' worth of data, then an attacker is simply going to ignore the one user who has a difficult to remove watermark.
That's precisely why I spend a little bit of time implementing those watermarks. I just assume that whatever entity that is holding my information will eventually leak it, so I might as well just try to use any kind of deterrent that I can. Other than that, I try to stay away as much as possible from KYC procedures but some people still haven't realized how much information they are giving away whenever they submit themselves to KYC...