Luca Fumarco

We identified 6 channels (mediators & institutional modifiers): Skill Acquisition, Self-Esteem, Parental/Teacher Investments, Selection into Programs, School Dropout Policies, and Delayed School Exit. There might be more! Specific behavioral mech. are understudied.

The Economics of Age at School Entry: Insights from Evidence and Methods (Forthcoming Article) - This article reviews the growing literature on age at school entry and its effects over the life course. Age at school entry affects a broad range of outcomes, including education, labor-market performance, health, social relationships, and family formation. We synthesize the evidence using a conceptual framework that distinguishes four empirically intertwined components of age at school entry: starting age, age at outcome, relative age, and time in school. Within this framework, we highlight six key channels through which age at school entry operates. While the effects of age at school entry are often substantial and persistent, many studies estimate bundled impacts without isolating specific components or directly measuring underlying mechanisms. We explain how different research designs capture distinct combinations of these components. We also highlight how institutional heterogeneity and behavioral responses can complicate the interpretation of results. We conclude by outlining directions for future research and policy design.

Forthcoming in the JEL: "The Economics of Age at School Entry: Insights from Evidence and Methods" by Mariagrazia Cavallo, Elizabeth Dhuey, Luca Fumarco, Levi Halewyck, and Simon ter Meulen.

#econsky

Working paper versions (online appendixes included) are available via, e.g., IZA, GLO, and EdWorkingPapers.

What ASE finding surprised you most? And where should future work focus — relative-age mechanisms? Long-run health? Family spillovers? Would love to hear. 9/9

For policymakers: ASE effects are real, multidimensional, and often long-lasting. And, we suspect, researchers are still just scratching the surface. 8/9

For researchers: the paper maps the literature and flags where conceptual and identification clarity can still improve. Appendixes cover 50+ paper summaries, school cutoff dates worldwide, and list 250+ readings. 7/9

Relative age is the most underappreciated component. Its effects are especially sticky in:
🔹 Behavioral assessments
🔹 ADHD diagnoses
🔹 Special education placement
Why? As a possible answer, schools benchmark kids against classmates — not against developmental norms. 6/9

A central insight: most ASE estimates are composite effects. Even clean RDD designs with grade-level outcomes bundle multiple components mechanically.
Knowing the exact source of variation is essential for results interpretation. 5/9

Methods matter — a lot. We walk through 2SLS, RDD, DiD, and reduced-form approaches, and clarify what each design actually identifies (and what it doesn't).
Cross-jurisdiction variation matters too — tracking systems, diagnostic rules, dropout policies — is often underused and underappreciated. 4/9

We propose a clean conceptual framework. ASE has four distinct components:
→ Starting-age (maturity at entry)
→ Age-at-outcome (how old you are when measured)
→ Relative-age (your rank within the class)
→ Time-in-school (cumulative exposure)
And they have an impact through various channels. 3/9

Being slightly older/younger when you start school matters. Age at school entry (ASE) shapes:
📚 Educational attainment
💼 Earnings & employment
🧠 Mental & physical health
👥 Social relationships
⚖️ Crime
💍 Family formation
Economists largely ignored this topic for decades. Not anymore. 2/9

🚨Forthcoming paper, @aeajournals.bsky.social 🚨#econsky
A thread 🧵 on "The Economics of Age at School Entry: Insights from Evidence and Methods". Humbled to work w/ @mariagraziacavallo.bsky.social, @bdhuey.bsky.social, Levi Halewyck & Simon ter Meulen 1/9

many thanks!

it is a survey of the literature on age at school entry. in some days, we ll post the usual thread w more details

yep, that's it (and bookmarklet is the word i should ve used instead of app, sorry)

you cannot install it on edge, or?

Grazie Alfredo!

Many thanks!

Aaah! Waking up in the morning to an acceptance decision from the Journal of Economic Literature...it feels great!

pitty altmetric app works only on a couple of browsers. great feature though

The political effects of X’s feed algorithm - Nature Among users initially on a chronological feed, 7 weeks of exposure to X’s algorithmic feed in 2023 shifted political attitudes and account-following behaviour in a more conservative direction compared...

www.nature.com/articles/s41...

yet again, a good reason for using this platform?

it's great: it reads like a 80s song title, econ vibes though

how was the murder rate in the years before covid-19?

Interesting paper on the gendered heckling the German Bundestag www.sciencedirect.com/science/arti...

hey #econsky, here is another cool AI: paperreview.ai from Stanford ML group. This is an AI reviewer, it (apparently) works well with econ papers! It is free, you just enter an API key, and then paperreview will email you a report within minutes.

hey #econsky, check this out: seminar-dry-run.vercel.app from @plausiblyexog.bsky.social. Very cool way to train for seminars: AI listens to your presentation, asks you questions, and then provides you with feedback!

I see some optimist in "new stable equilibrium"

Mediation analyses indicate that part of this effect operates through educational pathways: older students are less likely to experience delays and more likely to have student jobs. These factors explain some—but not all—of the relative age advantage.

Being almost one year older within a cohort increases the probability of employment one year after graduation by 3.5 p.p., and raises the likelihood of obtaining a permanent (+5.1 p.p.) or full‑time (+6.5 p.p.) contract.

Using rich Flemish survey data and a 2SLS strategy, we study whether being older than one’s classmates due to school entry rules affects the transition from school to work.