Comment by ComputerGuru

It's not throwing any information away because it can be faithfully reconstructed (via an admittedly arduous process), therefore no entropy has been lost (if you consider the sum of both "input bytes" and "knowledge of utf-8 encoding/decoding").