Actually, for a paramagnetic or diamagnetic material, there will be spin split bands, in the presence of an external field, B. The spin splitting, or Zeeman, energy is given by U=g\mu_B \mathbf{B \cdot S}, where g is the Lande g-factor. [1]
In a ferromagnet, you don't need an external field to see spin splitting. The exchange interaction between charge carriers (with spin s at position r) and magnetic moments (with spin Sn at positions Rn) is given by U_{ex} = -\sum_n J(|R_n-r|) \mathbf{s \cdot S_n} [2]. This can be calculated by your approximation of choice, and gives roughly, \Delta E = SJ(M), where M is the temperature dependent magnetization within a domain, and can be approximated by the Brillouin function [3].
Whether your material is ferromagnetic or not, the effect of the spin splitting on the position of the fermi surface is negligible[4]. However, depending on the temperature (and in the case of a para/dia-magnet, also the applied field) the spin splitting can result in a significant polarization of the Fermi surface (whereas, in the absense of spin-splitting, the Fermi surface would be expected to have an equal number of spin up and spin down electrons, and hence, no net polarization) [5].[1] See any solid state textbook : Kittel, Ashcroft & Mermin, Marder, etc. talk about this.
[2] This is the basis of RKKY theory. You can look for references on RKKY or Indirect Exchange.
[3] I've seen this is Kittel and Ashcroft. Note the difference between the J used here - for the exchange integral - and the J you will come across in calculations of the Brillouin function - for the total angular momentum.
[4] There are some exceptions, notably in quantum Hall systems, at low filling factors.
[5] This is the basis of Spintronics. A review paper discussing the use of diluted magnetic semiconductors for spintronics will likely talk about this.
If you're having a hard time finding references, I could look some up for you.
