As far as I am aware the 18% grey card was designed simply as a reference point, and was never designed to be used with exposure meters. It's one of those myths that seem to have come about over the years.
Myth or not. Reflected readings with the camera's spot meter from a 18% card always seem to match incident readings, and give correct exposures. Right or wrong, I will continue to carry on doing so. If it gave incorrect results I'd have questioned it myself some time ago. I've heard the arguments before, and done the maths before, but for the life of me I can not find any evidence that I'm under exposing a half stop.
This is not a new argument anyway, and for the life of me I can't work out why camera manufacturers don't just give clear information about how they calibrate their meters and just put the whole damned thing to bed. Nikon never disclose anything definitively, Sekonic, last time I read anything about this seem to go for 15%. I've been told in no uncertain terms from a technician from Leaf/Mamyia that they are DEFINITELY 18%.. LOL. I give up. You can't get a straight answer from anyone. All that article linked to above is explaining is the maths behind it all, but unless you know exactly what the camera manufactures are actually calibrating their meters to, it doesn't really help.
In reality... knowing that my Sekonic meters are calibrated to 15%, and assuming my Nikon could be 18% (if you believe Nikon USA) or 13% (if you believe Nikon Japan), then clearly the third stop difference isn't the big deal you all think it is... Which in itself, isn't that surprising either.
Having said all that, I bet if you sneaked into my camera bag and replaced my 18% card with a 12% card, I'd probably not notice unless I was being really critical.
I would do what I did years ago, and stopped worrying about it. I get great results with a 18% card, and have never paid any attention to Kodak's disclaimer. I'd probably get equally great results from a 12% card. The reality is though, I only ever use a grey card as an exposure aid in very difficult lighting conditions where an incident reading on location would not give me a clear indication of contrast ratios, so I spot off a grey card in various points on the scene. Given that I only ever do that in very difficult contrast scenarios I may just be so glad I got a good exposure that I don't examine it too closely. In controlled environments, and in pretty much 90% of location environments I'll be trusting my incident meter rather than ANY reflected reading.