Would a New Policy on Pot Be Good for the US?
Economists consider the societal impact of reclassifying marijuana.
Would a New Policy on Pot Be Good for the US?The US college-admissions process is notoriously burdensome for students, who have to research, visit, and apply to many schools in order to have a shot at selective institutions while still hedging against rejection. Colleges, meanwhile, waste time and other resources marketing to and sifting through many students who are unlikely to be admitted in the first place. Is this inefficiency inherent in the college-matching market? Chicago Booth’s Jacob Leshno and his coauthors decided to investigate, and suggest a procedure that could make the process easier and less wasteful for everyone.
(gentle music)
Narrator: For those looking to attend college in the US, the process comes with a lot of homework. College guidance counselors usually encourage applying to six colleges or more to hedge against rejections. From researching colleges to taking standardized tests, writing essays and soliciting recommendation letters, the entire process can take a hundred hours or more. Throw in application fees and campus visits, and finding the right college can be quite expensive. The most competitive institutions accept just a small fraction of the students who apply. Stanford, for instance, accepts only around 8 percent of applicants. It raises the question: Could this process be more effective for students and for the universities too? Chicago Booth’s Jacob Leshno and his coauthors have developed a model to examine the challenges and see if there’s a better way.
Jacob Leshno: So we wanted to think about the whole problem that prospective college students face in finding out where they want to go to school. You probably need to visit or read about them or understand them to understand whether you want to go to one program or another. And before you spend all this effort understanding college programs, you first need to understand which college programs will actually accept you. There’s no point in going to travel to figure out whether you like a college if there’s no chance that the college will accept you. But that’s also one of the sole points in the process, that I have this range of schools that I’m uncertain about and I need to maybe visit more colleges than the one that I think is the most relevant for me because I don’t know whether I for sure will get in or not. And that got us started thinking as theorists: Is this some essential feature of this market or is this just like a bad property of the American education system that there’s all this uncertainty? Can we reduce this uncertainty? Is this uncertainty inherent to the market? We wanted a model that will tell us that.
Narrator: If students had better information about their admission chances, they would spend less time visiting and inspecting colleges and apply to fewer of them. This shift would save the students time and money and would likely benefit colleges too. Schools could save money on marketing by targeting students they actually intend to admit, and their admissions officers could spend less time going through applications that they’re ultimately going to reject. But is it possible to determine in advance which colleges will accept which students? The short answer is not exactly. To understand why this is a tough question to answer, suppose for now that every student has decided their college preferences, ranking the schools in which they would like to attend. At the same time, each school could create a ranking of students using a set of criteria it chooses. It could give a score to each student and rank order them.
Jacob Leshno: So in economic terms, or in the language of the model, the outcome that we’re looking for is an assignment where each college admits the students it considers to be the best, the ones it gave the highest score. And students accept an admission offer, filling the colleges, and all students end up taking up the admission offer to their most preferred college. If it was possible to calculate and publish each college’s admission cutoff, students would know before applying which college would accept them. Then they could more easily determine which schools they like the most and apply only to schools that they like the most and would accept them. But schools can’t rank students and create cutoffs until students apply and decide which schools they prefer. In the model, already we get that there’s an impossibility because students will wanna know those thresholds, those admission cutoffs to know where they can get in and form the preferences. And they can’t really tell you those preferences before they get the thresholds. But colleges cannot calculate the thresholds before they know student preferences, because before they know student preferences, they don’t know how selective they can be—which students will go to them and which students will go somewhere else. So this is actually a chicken-and-egg problem, and it’s theoretically impossible to solve.
Narrator: However, Leshno and his coauthors show that this problem can be particularly avoided by turning to historical data. It’s impossible to publish exact cutoffs, but it is possible to estimate these cutoffs by using data from previous years. And they find through modeling that publishing colleges’ estimated admissions cutoffs could still make the process easier for the vast majority of students.
Jacob Leshno: That means that most students can inspect schools before their admission season even started. Before you apply anywhere, you can essentially know where you’re gonna get in. Of course, there’s gonna be some students that are gonna be on the margin. It’s hard to predict. But for most students, you know which programs will accept you, and now you have a much easier task at hand. You just need to figure out which one you want the best. That allows them to search with minimal hassle and find the best program for them. So we point out that this is not just a good outcome in our model. It’s what’s done all around the world in various form. In most countries, college admission is calculated to national exam scores, and schools like to tell the applicants in advance what are the scores they need to get into particular programs. A lot of colleges put a lot of investment into that to really inform candidates whether they can get in or not. And while college cutoffs fluctuate from year to year, it’s still possible to get sufficiently good information to inform more students of which program they are likely to get into.
Narrator: If US colleges were to adopt these more transparent practices, they could eliminate some of the inefficiencies and incentive problems that currently make the process so difficult. This could make matching students with their best-fit colleges easier and more efficient for everyone.
Economists consider the societal impact of reclassifying marijuana.
Would a New Policy on Pot Be Good for the US?Chicago Booth’s Chad Syverson discusses the role of good management in the age of AI.
Why AI Might Not Make You More ProductiveMonetary policy decisions can have meaningful long-term effects.
How High Interest Rates Harm InnovationYour Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.