Normally when people talk about numerical bases, it is said that they are entirely equivalent. However, is it not true that the lower the radix, the greater is the information density that is immediately conveyed? For example the binary 11111 is equivalent to the decimal 31, but the former provides more immediate information. Transforming (encoding or decoding) between these systems then necessarily requires an algorithmic computation.
If this is correct, what are the immediate ramifications?
In binary representation, each symbol can represent two values; in information theory terms, that amounts to $1$ bit per symbol. In decimal representation, each symbol can represent ten values; that amounts to $\log_2(10)=3.3219$ bits per symbol. So, yes, decimal representation has more information density ($3.32$ times), and hence it's more efficient (in some sense). And hexadecimal representation has $4$ times more.
I'm not sure what "immediate ramifications" you are thinking of.