The CJK languages are syllabic languages. And each syllable occupies two bytes in computer memory. There is no such thing as a single letter in the CJK languages.
The Chinese, Japanese and Korean (CJK) languages are character-based with each character referring to an idea - as opposed to a specific shape of the character or an object. Because their characters are more complex and graphical than Roman alphabet letters, they typically use twice as much memory and are considered double-byte languages.
The complexities of the double-byte languages raises issues of file size, operating system and software compatibility, and even the clarity and readability of text due to fewer font choices. And, because character text is semantic-based, a phrase cannot simply be broken at any random point.