If I may add a few words...May sound boring
Theoreticaly speaking a light source's temperature description of colour should match the black body radiation at the given temperature...for example, a bulb with a 6400K spectrum should emit a light curve of a black body, heated at 6400K.
However, cfl's are not black bodies, they are basically electrically excited gassous tubes in which the process of light emission is very different from the black body(heat radiation). Because of that, the tubes are filled with coloring chemicals which control the proportion of the most prominent spikes in the mercury spectrum and work as a kind of filter to the emitted light. This means the overall color of the lamp is a clever proportion of some specific wavelenghts of light in order to match the human eye's perception of the supposed black body radiating.
What's the point of that? I think it's the lack of "supposedly-be-there" wavelenghts because of the nature of gas light emission. The temp color lables are fundamentaly deceptive in cfl world, but handy because we deal with light with our own eyes, so the eyes can be tricked without realizing their own limitations
.....
I measured some cfl's spectrums and found out that the 2700K lamps have veeery little blue light and are more than 6x less adequate for vegetative growth(6x less blue light compared to 6400K). The 6400K however, has little less red, but overall should be more suitable for flowering than the 2700K's are for vegging.
Based on that observation I would always switch lights when switching cycles and would probably look for higher than 6400K color. A mix of 6400K and 8000K* would be of benefit during veg grow, IMO.
*not sure really, would have to measure