The past seven years have seen a huge increase in the efficiency of perovskite based solar cells going from 3.8% to 22.1%. The 2.2 eV bandgap of methylammonium lead bromide (CH3NH3PbBr3) makes this material a possible candidate to serve as a top cell in a tandem solar cell, combined with a crystalline Si bottom cell. To realise this, it is necessary to understand the light-induced charge carrier dynamics in this perovskite, specifically the charge carrier lifetime and mobility. CH3NH3PbBr3 has three phases with transitions from cubic to tetragonal at T = 236.9K and to orthorhombic at T = 144.5K. We chose single crystals to study the intrinsic properties of CH3NH3PbBr3, since there are limited number of grain boundaries and the crystals are phase pure. These crystals were studied by temperature dependent complementary time-resolved microwave conductivity (TRMC) and photoluminescence (PL) measurements. In the cubic phase, TRMC results show on excitation at 560 nm a mono exponential decay with 250 ns lifetime which is probably due to Shockley Read Hall recombination. On excitation at 500 nm, which leads to generation of charge carriers close to the surface, a much faster multi-exponential decay is observed. This faster decay might originate from intraband-gap surface states that lead to fast recombination. Furthermore, the yield of charges decreases on higher excitation densities. This reduction in yield, which is in favour of exciton formation in agreement with the Saha equation. The intensity independent PL lifetime measurements with lifetimes in the order of several nano seconds are in accordance with above view. Interestingly, in the orthorhombic phase apart from excitonic emission at 560 nm, a broad second emission at 620 nm (DE = 0.2 eV) was observed. The TRPL of this 620 nm peak was characterised by a slow rise of several tens of ns and a decay extending into the microsecond regime. This slow PL decay corresponds with the TRMC trace, implying that the PL originates from the radiative decay of mobile charges. We attribute the slow rise to carrier diffusion followed by localisation at domain boundaries. Rashba splitting of the valence band due to symmetry breaking at the boundaries can explain the low energy emissive states. These boundaries might be related to the differences in crystal orientation in the orthorhombic phase. On heating the crystal back to the tetragonal phase leads again to disappearance of this low energy PL.