The release of institution-level earnings information as part of the Obama Administration's new College Scorecard data has already spawned new "value-added" rankings based on the economic outcomes of students who attended similar institutions. These emerging rankings are an improvement on simple unadjusted rankings, but the wide variance within and among these rankings prompt questions about the precision of this methodology. This report proposes a value-added model that parallels a type of value-added model commonly used in K-12 education--one that examines the same metric over time. Just as K-12 value-added models use growth in student achievement to assess teacher impacts, the most compelling value-added model for higher education institutions tracks student earnings from prior to enrollment to after leaving college. The authors demonstrate the feasibility of this method by calculating an economic mobility indicator for two groups of students: the mostly older students who were independent for federal financial aid purposes, and the younger students who filed as dependents. Findings reveal substantial variation in economic mobility across colleges, including among colleges where students come from similar socioeconomic backgrounds. An economic mobility ranking provides a more conceptually streamlined approach than many existing ranking methods, but doing it well requires more detailed data than are currently available. The most important limitations of current data are: (1) they report family income prior to college but individual data after; and (2) they report income prior to college for all federal aid recipients but only report earnings after college for former students who are employed. The authors recommend that the next release of College Scorecard data facilitate economic mobility rankings of colleges by reporting more detailed earnings breakdowns that address these limitations.