Research

with Nagisa Tadjfar

Abstract: Elite university access is highly unequal. Low-income students are less likely to apply to and attend than equally qualified high-income peers. Using UK administrative data, we exploit “breakthrough” events when a school first sends a student to a top university. Applications from that school to that university subsequently rise by 30%. Students induced into elite universities by a breakthrough are lower-income, but graduate at typical rates. Access induced by breakthroughs promotes upward mobility: marginal entrants earn £4,000 more annually than matched control students. Why were these students not applying previously? Using a field experiment in British schools, we show that the primary barrier is students' beliefs about social fit at top universities rather than beliefs about admissions chances or success at university. At baseline, low-income students are more pessimistic about their chances of fitting in at an elite university, but not about their chances of admission or graduation. Students randomly assigned to view short videos of undergraduates discussing their experiences are 6 percentage points more likely to apply to the speaker's university. While students’ expectations of fitting in and making friends shift, beliefs about admission chances or graduation do not. Students randomly matched with mentors primarily discuss social life at university, and the most important factor participants raise with mentors is whether they would fit in and enjoy their time. Our findings highlight perceptions of the social environment at elite universities as a central barrier to access, and illustrate that scalable interventions can address this barrier and promote social mobility. 

with Nagisa Tadjfar

Abstract: The use of screening algorithms such as standardized testing in university admissions is widely criticized for benefiting wealthy students and not accurately capturing the potential of low-income students. Does standardized testing inhibit opportunity? Our paper answers this question using a staggered elimination of pre-university testing requirements in favor of teacher-assigned grades in the UK. First, we find that low-income students become 3pp more likely to attend university, while enrollment among high-income students is unchanged. However, only 1 in 3 of these marginal students graduates on time. Despite this low graduation rate, students shifted into university see large private returns: they begin their careers at better firms and gain £50k—£100k in lifetime earnings, net of tuition fees. Second, we see upstream shifts beyond the university enrollment margin — affected low-income students are more likely to take academic-track subjects in high school and apply to university. Third, we show that standardized tests in our setting exhibit no calibration bias against low-income students, whereas teacher grades favor them. Taken together, our results indicate that switching from tests to teacher grades can expand opportunity for disadvantaged students even in the absence of calibration bias in tests. Our findings suggest that reducing testing barriers expanded educational investment earlier in the pipeline and delivered long-run gains for marginal entrants.  

with Phi Adajar and Nagisa Tadjfar

The mid-2010s saw a rapid proliferation of offers to UK universities in which students were admitted regardless of their end-of-school test results, effectively shifting risk from students to universities. In this paper, we seek to understand the nature of this unraveling, with a focus towards universities’ incentives and the effects on students’ short- and long-term welfare using data from all UK college applications from 2012 to 2021. We find that these “unconditional offers” were given by lower-ranked universities and targeted towards higher-achieving students. Students with an unconditional offer are 4.5pp less likely to attend their top offer. This is consistent with students being shifted into lower-ranking universities and becoming “undermatched”. Simultaneously, college attendance increases; students with an unconditional offer are 12.4pp more likely to be matched to a college when compared to students with similar test scores and teacher evaluations. On the university side, these unconditional offers improve university yields and student composition, and correlational evidence shows that a university’s adoption of unconditional offers occurs in response to its competitors doing the same. We estimate a structural model of university offers to evaluate the nature of this competitive response and the implications for student welfare. 

with Taha Choukhmane, Cormac O’Dea, Jonathan Rothbaum, and Lawrence Schmidt

IFS Briefing Note BN234, with Barra Roantree

PDF