Financial aid policies are widely used to foster access to higher education for low-income and underrepresented students. Prior research has documented the positive impacts of these programs on the outcomes of the students it intents to help (e.g., Mello, 2021; Londono-Velez et al., 2020; Bleemer, 2021a; Black, Denning, & Rothstein, 2020). However, how the influx in the number of low-income and underrepresented students at selective institutions shape student outcomes is a somewhat overlooked aspect. Increasing the number of low-income and underrepresented students may come at the cost of reducing the overall academic performance of the group (Arcidiacono et al., 2015). Moreover, it could also lead to social segregation between wealthy and low--income and underrepresented students within the institution. This is an undesirable outcome, if we take into account the increasing evidence on the importance of social membership and networks for employment and overall social mobility (Marmaros & Sacerdote, 2002; S. D. Zimmerman, 2019; Michelman et al., 2021; Rivera, 2016). This dissertation starts by addressing two questions. First, what are the consequences of increasing the presence of low-income students at an elite college on students' academic achievement? And second, can this form of desegregation lead to more interactions between low-income and traditionally privileged students? To answers these questions, I focus on an Elite University in Colombia which experienced a large influx in its enrollment of low-income students, due to the implementation of a large financial aid program known as "Ser Pilo Paga" (SPP). The program induced plausibly random variation in the shares of low-income students within cohorts and across majors, which I use to examine how the changes in peers' composition impacted the traditionally privileged students attending this elite institution. In the first chapter of this dissertation, I use administrative data from the Elite University and document how the influx in the share of low-income students led to significant achievement gaps between wealthy and low-income students. Then, I use a difference-in-difference research design to examine the effect that this influx in low-income peers had on wealthy students academic performance and persistence. Overall, my results indicate the influx in low-income peers had no impact on the academic performance of the students traditionally attending this institution. Peer effects estimations suggest that the relatively low academic performance of low-income students had no impact on wealthy students. These findings complement those from Bleemer (2021a), who find college re-segregation has no impacts on the academic performance of White and Asian students. My results are also consistent with K-12 evidence showing desegregation policies have no impact on the students traditionally attending the desegregated schools (Angrist & Lang, 2004). In the second chapter, I examine whether the influx in the share of low-income peers at the Elite University led students to diversify their social interactions. I complement the administrative data and research design from the previous chapter with administrative records on students' co- movements across campus captured by turnstiles located at all entrances. To validate the turnstile-elicited interactions, I use secondary survey data on social networks from a sub-sample of students at Elite University. The 9.5 percentage points average increase in the share of low-income peers at an entry cohort--major led wealthy students to double their connections with low-income peers. At least half of the increase in interactions between wealthy and low--income students, however, is explained by interactions of wealthy students with low-income but high-achieving students. These results suggest students diversify their interactions primarily among students with similar academic achievement levels. In the last chapter, I focus on a broader aspect of financial aid policies, and examine how student loans impact college and employment outcomes. In this chapter, my co-author and I are particularly interested in understanding the monetary returns to increasing investments in college education through student loans. To do so, we focus on a nation-wide student loan program in Colombia known as ACCES that covered 75 percent of the tuition cost for low-income students, and exploit rich administrative data to assess the impact of the loan. Importantly, the cost of tuition per semester is fixed for full-time students, regardless of the number of credits they take. The eligibility to the loan increased college enrollment by 9.6 percentage points, the number of semesters attained in 1.76 semesters, and college graduation in 15.4 percentage points. We also find evidence suggesting students with the loan had salaries 13 percent higher six years after high school graduation. To shed light on how student loans can lead to better salaries, we estimate the returns to additional college investments as captured by the marginal increase in college semesters found among loan recipients. We estimate a return to the additional college semester of 3.5 percent of the average daily salary. Sub-group analyses suggest much of the return to the additional semester of college is driven by college completion. This chapter contributes to the literature that has examined how relaxing credit constrains can have positive impacts on long-term outcomes by increasing human capital investments (e.g., Black, Denning, Dettling, et al., 2020; Cameron & Taber, 2004). [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]