Tigger is correct. The theoretical maximum bit rate can be calculated by solving for R from Eb/No = SW/NR where Eb is the energy per bit, No is the noise power per bit, S is the signal strength, W is the bandwidth, N is the noise power spectral density, and R is the bit rate. In practice, as Tigger also notes, it's a lot more complicated! For example, in a noisy or low signal environment, you might use error detecting/error correcting encoding (i.e. Viterbi or Reed/Solomon) to get a lower bit error rate, but you would then transmit more bits (the encoding bits) than you would for unencoded -- thus a lower effective bit rate. Entire books are devoted to this subject. Sklar's is one of the best, see first reference.
To simplistically answer your question, power received is related to the gain of the receiving antenna, the power transmitted, the gain of the transmitting antenna, the distance, the wavelength, and the gain of the receiving antenna. See equation 5.13 in second reference.
Pr = (PtGt/4piR^2)(lamda^2/4pi)Gr
In your example, all are constant except R^2. So received power decreases by the square of the distance. Since bit error rate is determined by Eb/No, insist that Eb/No, in the first equation, be held constant. Thus, your bit rate also decreases by the square of the distance.