On Gopalsamy and Liu's conjecture for global stability in a population model