You can store Unicode data in SQL Server today, but only in UCS-2 encoding. In many environments, UTF-8 is the dominating encoding for Unicode data, and it is the default
encoding for XML documents.
Today UTF-8 applications must convert between UCS-2 and UTF-8 to retrieve and store
data in SQL Server, which adds complexity and overhead to applications.
Another issue is that for languages written in the Latin script, UTF-8 is a much more effective storage format, since most for characters in text, only one byte is needed.