Due to the recent economic recession, California has cut billions of dollars in general fund support for higher education (Center for the Study of Education Policy, 2012; Taylor, 2012). These cuts have increased tuition and restricted access to California’s four-year institutions, pushing more students into the community college system (Clark, 2009; Varlotta, 2010). The combination of decreased resources and increased enrollments have resulted in oversubscribed courses, and a recent survey by the Pearson Foundation found that California’s community college students were almost twice as likely to report being unable to enroll in courses than the national average (Johnson, 2013; Pearson Foundation, 2011). California has taken steps to restrict community college access by calling for students with high unit counts or lower academic performance to be placed in the back of the registration line, but it is unclear how changes to the law will impact students as little work has examined how registration systems influence student behavior (California Community Colleges Student Success Task Force, 2012; Little Hoover Commission, 2012; Taylor, 2011). Some evidence suggests that when colleges are unable to increase supply to accommodate demand that resources per student are reduced, university quality may decrease, and students’ degree completion rates decline (Bound, Lovenheim, & Turner, 2010; Kane & Orszag, 2003), although this may be true in four-year institutions more than in community colleges (Bound, Lovenheim, & Turner, 2012).

This paper utilizes detailed micro-level data from a large California community college to examine whether students with later registration dates exhibit different academic outcomes. I discover a number of important student and institutional behaviors that provide insight into overcrowding. Most importantly, by directly assessing the exact time at which each course section closes, I find that there are a significant number of open seats available to almost all continuing students and a large number of new students when they are first eligible to register. This paper also identifies a number of issues that illustrate the complexities of student enrollment: few students use their registration time effectively, waiting too long to register; many students who successfully register in a course either drop or withdraw from the section; and most students closed out of general education courses do not attempt to register in alternative but equivalent courses. As a result of these behaviors, I find two different kinds of “new” students. About one-third of new students matriculate on-time, are assigned a registration block that gives them access to many key courses, and attempt a high number of units that is comparable to continuing students with earlier registration times. In contrast, two-thirds of new students matriculate close to the beginning of the semester, rarely take math or English placement exams, attempt few units, and are much more likely to leave the college after one semester. With these findings in mind, I estimate that pushing high-unit and low-academic students to the back of the registration priority line will open approximately 4% to 6% of seats in the most heavily impacted courses.