Grades are important for admission of students in most higher education programs. Analyzing admission and student performance data at a major Norwegian business school, we find that the grading practice of teachers at regional colleges sending students to the school is affected by the average performance of the students being graded. Teachers at colleges recruiting good students from upper secondary school tend to be strict in their grading practice, while teachers at colleges recruiting less good students tend to follow a lenient practice. This has implications for the interpretation of grades and hence for optimal admission procedures. We develop a methodology to assess the consequences of differential grading standards. Approximately 10% of the students in our data are admitted at the expense of more competent students. We demonstrate costs for the school admitting the wrong students and in particular for the rejected students. (Contains 6 tables and 9 footnotes.)