We have been down this road before. It is all about the
power triangle.
They quote the phase angle between current and voltage at two frequencies: -
Phase (@85 / 4200 Hz) -16 / +25 degrees
The reason for just specifying those frequency points is not stated. It would be nice if they had shown a graph with current and voltage plotted against frequency.
However at 85 Hz there is a phase angle of -16 degrees. That tell us that the load is largely capacitative at that point whereas at 4200 Hz it is +25 and largely inductive.
Where impedance and phase come out is the luck of the draw. You basically design for a flat frequency response with minimum violence to time and phase and you take what you get.
I really hope we can move away from the passive crossover. These issues don't apply with active designs.