Oh, no, it's doing something. It just doesn't affect the contrast in the final image the way you think.
I keep having to explain this, because those labels in enfuse mean something quite different than in other tools.
Enfuse doesn't adjust colors at all, unlike tonemapping. That's why the default results tend to be much more naturalistic than with HDR/tonemapping. Whatever pixel values are there in the original member images are what you're going to get in the final image. It's just that individual pixels may come from any one of the member images.
The three selection criteria are based upon contrast, saturation, and exposure, and messing with a slider labelled one of those three things in an enfuse front end means you're adjusting the importance (weight) that factor has in selection. If you weight most heavily on exposure, you're exposure-fusing. If you weight most heavily on contrast, you're focus-stacking. If you weight most heavily on saturation, you're picking the most saturated pixels among the member images.
This is why I mentioned I do enfuse, and
then I do a curves adjustment. Most HDR/exposure fusing methods in essence are doing the equivalent of a reverse-S curve adjustment or lowering contrast, so by default a lot of hdr image placed into a smaller range somehow are going to look flatter coming out of the algorithm. In fact, lowering contrast with in-camera processing is how some video shooter extend dynamic range.