Thank you for clarifying. Can you comment on the relative expressive power of Shen's type system versus e.g. Haskell? I understand that the former can accept as

Thank you for clarifying. Can you comment on the relative expressive power of Shen's type system versus e.g. Haskell? I understand that the former can accept as well-typed terms the other can't; in what way is this a consequence of modeling Shen's type system using the Sequent Calculus versus Hindley-Milner deduction?