As usual, the answer is not so clear.
There is not such thing as negative fluorescence. The negative fluorescent component is an artifact of the was fluorescence is measured according to:
ISO 2469 Paper, board and pulps - Measurement of diffuse radiance factor
ISO 2470-1 Paper, board and pulps - Measurement of diffuse blue reflectance factor - Part 1: Indoor daylight conditions (ISO brightness)
ISO 2470-2 Paper, board and pulps - Measurement of diffuse blue reflectance factor - Part 2: Outdoor daylight conditions (D65 brightness)
This is a step-by-step process. First, brightness is measured with the appropriate UV Level for the situation, C or D65. Then a UV-cutoff filter is introduced and brightness is measured again. Here is the data:
Why is this? The cutoff filter is active at 420 nm. Therefore, data below 420 is essentially not available in the UV-EX condition, The recommended procedure for handling this is to simply replace the values below the 420 nm cutoff with the same reflectance value. This can be seen in the following graph.
Numerically, it looks like this...
Since the reflectance values from 400 to 500 nm wavelengths are used in the calculation of brightness, it yields a negative fluorescent component. So, back to the original question, is a negative fluorescent component real? According to the directives of ISO 2469, ISO 2470-1 and ISO 2470-2, the fluorescent component is negative and should not be zero. However, just as there can be some intrinsic fluorescence (from the raw fibers without any chemicals added), there can be a small negative value due to the procedure used to measure fluorescent component. Therefore, when fluorescent component is in the rang of +/-0.25, it is effectively zero.