Students who attend different colleges in the United States end up with vastly different educational and labor market outcomes. We estimate value-added of individual postsecondary institutions to disentangle causal impacts of colleges from student sorting in producing these disparate outcomes. Linking administrative registries of high school records, college applications, admissions decisions, enrollment spells, degree completions, and quarterly earnings spanning the Texas population, we identify college value-added across the diverse distribution of thirty Texas public universities by comparing the outcomes of students who apply to and are admitted by the same set of institutions, as this approach strikingly balances student ability measures across college treatments and delivers value-added estimates impervious to additional controls. We find that differences in causal value-added play a much smaller role, relative to student sorting, in producing observed outcome differences across colleges. The distribution of value-added is not degenerate, however, and while it has little relationship with selectivity, we find that non-peer college inputs like instructional spending and the faculty-student ratio do covary positively with value-added, especially conditional on selectivity. Examining potential mechanisms, colleges that ultimately boost earnings also tend to boost persistence, BA completion, and STEM degrees along the way. Finally, we probe the potential for (mis)match effects by allowing value-added to vary flexibly by student characteristics, including race, gender, family income, and pre-college measures of cognitive and non-cognitive skills. At first glance, black students appear to face small negative returns to attending more selective colleges, but this pattern of modest “mismatch” is driven by two large historically black universities in Texas that have low selectivity but above-average value-added. Across the non-HBCUs, black students face similar returns to selectivity as their peers from other backgrounds.