Published 26 August 2022 by Benjamin Skuse
Going Micro for a Better World
Whether you develop theories for big-picture macroeconomic forces or focus on developing hypotheses around examples of individual economic decision making, data are the bridge linking what you think happens with what really is happening in the real world.
In recent decades, the increasing use of large individual‐level data sources to analyse economic behaviour alongside major advances in econometrics – which uses economic theory, mathematics and statistical inference to quantify economic phenomena – have strengthened this bridge for microeconomics, leading to a revolution that has allowed unprecedented insights into the role of personal and business decisions in evaluating potential outcomes. These insights are increasingly being sought by decision- and policymakers on a host of different and important topics.
In his lecture during the 7th Lindau Meeting on Economic Sciences on Wednesday, 24 August, titled ‘Designing Financial Aid for Research’, Joshua Angrist provided an impactful example of how microeconomics and the methods he developed to analyse causal relationships in natural experiments that earned him the 2021 Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel (alongside David Card and Guido Imbens) can be wielded to inform important policy decisions.
It is well known that the US Government and private organisations fork out billions of dollars in financial aid to undergraduates in the form of subsidised loans and grants. “The public and policymakers hope that this is money well spent, meaning that the aid is causing people to get more schooling than they otherwise would have,” said Angrist. “So you’d like to know whether the aid is actually boosting post-secondary education: that’s a great causal question.”
Angrist sought to answer this question when, in 2011, he and colleagues partnered with the Susan Thompson Buffett Foundation (STBF), one of the largest private funders of college scholarships in the US. This partnership allowed Angrist to implement a partially randomised study of financial aid through the targeted allocation of STBF scholarships covering the full cost of tuition and fees to Nebraska high school graduates.
Using STBF’s applicant scoring system based on means, merit, need and other factors, Angrist and colleagues focused on applicants he describes as “on the bubble”, so that the most and least deserving remained unaffected: “Thumbs up on the right, thumbs down on the left, and then in the middle, we toss a coin”. Any differences in outcomes between the winners and losers in this middle group reflect the causal effect of the STBF scholarship.
Following these study subjects for six years (and now continuing to follow them into adulthood), Angrist and colleagues have already revealed interesting trends. Unsurprisingly, those receiving scholarships have higher bachelor’s completion rates than the control group, by about eight percentage points. “It’s not revolutionary, but we’re used to this in social science,” said Angrist. “Most things that you try to do have no effect. So anytime you move the needle, you’re quite pleased.”
More interesting is that financial aid appears to have more impact on students who are less ready for college. “It turns out that the lower your test scores, the bigger the effect of the aid,” said Angrist who calls this effect the ‘merit aid paradox’: “In general, donors and public policy in this domain tends to favour people who have high GPAs and high test scores, but that’s actually counterproductive from the point of view of increasing college attendance.”
Another surprising finding is that the impact of financial aid heavily depends on the length of course a student applies for and when aid is given. “Paying for somebody’s community college degree does not increase the likelihood that they get a community college degree,” Angrist revealed. “So that’s an interesting and maybe discouraging finding for people who think it’s very important to subsidise two-year programmes.”
For all groups studied, optimal outcomes come from targeting aid early to students enrolling in four-year courses. “Aid works exclusively through this early engagement channel. In fact, we call that an exclusion restriction, and the strong causal claim that I’m making is outside of this channel, financial aid is irrelevant,” Angrist argued. “Maybe we should think about spending the money a little bit more efficiently by front loading it, thinking about ways to give people money that gets them into four-year programmes, but spending less as they work their way through the programme.”
The Long Reach of Microeconomics
Angrist’s Lecture focused on just one of a diverse range of applications of his microeconomics advances. For example, he also applied his Nobel-winning methodological work to predict how military service affects lifetime earnings, and how the causal effects of education on wages are related to season of birth.
Today, economists are applying microeconomics methods to a plethora of specific and important problems. No where has this been more clearly demonstrated than in the series of three ‘Next Gen Economic: Applied Microeconomics’ sessions at #LINOecon. In each rapid-fire session, 10 young economists had six minutes each to present their exciting work interrogating economic questions concerning Environment, Health and Labour on Thursday 25 August, and in Environment, Development and Labour, and Political Economics, Gender and IO (input-output) on Friday 26 August.
Attendees were given a break to refresh themselves halfway through each marathon two-hour seminar, and for good reason. In the first session, audience members were taken on a whirlwind trip around the world, one minute hearing about how an intervention to match immigrants with members of the resident population led to improved immigrant labour market outcomes in Sweden, the next minute learning of an investigation on how fear of radiation exposure from the Fukushima disaster in expectant mothers has impacted birth outcomes.
The two parallel sessions on Friday followed a similar course, charting the impacts of international, national and local interventions on society, culture, the environment and the economy. Everything from the political impact of Greta Thunberg’s Fridays for Future movement to the geography of black economic progress after slavery was interrogated through the lens of microeconomics.
Building a Critical Mass of Micro Studies
This eclectic mix of microeconomics applications was reflected in the backgrounds of the panellists for Wednesday 24 August Panel Discussion ‘Applied Micro Revolution’. Speakers included Angrist; fellow Laureate Daniel McFadden who, for example, applied his theoretical discoveries to determine demand for the San Francisco Bay Area Rapid Transport (BART) system; young scientist Anne Sophie Lassen, who studies the mechanisms behind and implications of gender differences in labour market choices (and also featured in Applied Microeconomics: Political Economics, Gender, IO); and young scientist Michael Pollmann, who has analysed the causal effects of grocery stores on foot traffic to nearby businesses during COVID-19 shelter-in-place policies.
Taken individually, it could be argued that each of these applications (and indeed, many of the applications outlined in the three Applied Microeconomics sessions) appears niche. On this point, an important question came from the floor: are micro studies so focused on tackling very specific questions in a causal way that they have little broader relevance?
“I don’t really agree with that,” argued Angrist. “What are my students working on? They’re working on things related to public policy in the labour market, minimum wages, trade, levels of competition in labour markets, monopsony, the gig economy – these questions seem no less important.”
McFadden agreed and took this argument further, suggesting that the accumulation of answers to specific questions in microeconomics is in itself slowly driving the entire field forward. He sees a grand theory of microeconomics and microeconometrics around the corner.
“In other sciences, we think of the theory of evolution or the theory of relativity as something that sprung in whole from the minds of brilliant scholars. But in fact, in each case, they were the result of years of accumulation of results that were nibbling at the edges of other previous theories showing that there was something wrong with them,” McFadden explained. “Now, you could say that each of those little studies was a small piece, not so much important in itself, but the cumulative effect was to create the environment where a grand theory of what was going on could evolve… I think we’re at the same stage that the biologists were when they were busy classifying bugs in the species – we’re just waiting for a Darwin.”