According to this link sound (especially high frequency sound) is more attenuated in fog, because it is dispersed by the (billions of) air-water interfaces of all the droplets. This is one reason why a fog horn is a very low sound - low frequencies travel further, especially in fog. For echolocation you want to use high frequencies, and fog is more attenuating at those frequencies. Another site confirming this information is this one - slightly older than the other one. Of course one might wonder whether one article got its information from the other (it does look similar...).
Finally there is a very thorough (if old - 1953) paper on the attenuation of sound in fog in the Caltech library. Interesting diagram from this paper confirms that the attenuation of sound in fog is stronger at higher frequencies (for details of the definition of β see the paper):
Even with the question of attenuation - when visibility is reduced to 40 ft, echolocation will probably beat it handsomely at intermediate distances. There are a couple of other interesting things you can do to improve your ability to see in fog.
1) Yellow "driving glasses". These work because they cut out the blue components of light. When fog droplets are very small, light scattering is in the Rayleigh regime - that is, scatter probability goes as the inverse fourth power of the wavelength, and blue (400 nm) light is 16x more scattered than red (800 nm) [note - using round numbers...]. By cutting out the blue component, you reduce the amount of scatter that reaches the eye and improve the contrast. Skiers also use yellow "fog glasses".
2) Scanning light source. This is one of those magical things that ought not to work but does. With normal (flood) illumination, light scatters "from everywhere to everywhere". If instead you look along the line of (say) a laser shining into the fog, then the only scattered light you see is the light that scatters exactly 180 degrees back at you - which is a small fraction of all the scattered light. If you scan the light source and detection system in sync, and very quickly, you can build up an "almost scatter free" image. This raster scanning technology is used in some underwater search applications and can penetrate about 6 "attenuation lengths". As was discussed in the comments, this method actually works best when the viewing angle is not exactly 180 degrees - not only is the back scatter from the fog weaker (there is a curious doubling of scatter intensity that happens at exactly 180) but also, by looking at a slight angle, you are able to eliminate the back scatter from the closest fog - greatly improving penetration.
More recently researchers in Israel have come up with a way to image through thin layers of scattering material - as you can see in the link, they are also able to see "through fog" (although it was not clear to me whether their technique can apply to actual imaging in fog).