Estimating the Technology of Children's Skill Formation - Working Paper 22442

Sep 18, 2026

Uploaded by Grenda

Estimating the Technology of Children's Skill Formation studies how children’s skills develop and how parental inputs translate into later outcomes. The paper uses a dynamic latent factor structure to separate latent skill development from measurement error, showing that constraints on observed measures support identification of skill-technology features. It then estimates a joint dynamic process for latent investment and skill accumulation, allowing static and dynamic complementarities. Using U.S. data, results highlight strong early-cognitive productivity of parental investments for ages 5–6, especially for children starting with lower skills, supporting targeted interventions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages87

Reading Time3 h 39 min

AccessFree to download

Introduction

Measurement issues and identification challenges

Contributions: measurement restrictions and dynamic skill-formation model

Age-invariant measures and their implications

NBER WORKING PAPER SERIES ESTIMATING THE TECHNOLOGY OF CHILDREN'S SKILL FORMATION Francesco Agostinelli Matthew Wiswall Working Paper 22442 [http://www.nber.org/papers/w22442](http://www.nber.org/papers/w22442) NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 July 2016, Revised March 2022 We thank the Editor, Jim Heckman, and four anonymous referees for their comments and feedback. We also thank Lance Lochner, Morteza Saharkhiz, Greg Veramendi, and participants at seminars and conferences at Arizona State, Barcelona GSE Summer Forum, Brown, CEHD (UChicago), Collegio Carlo Alberto, Duke, Interaction Workshop 2018 (Northwestern), Minnesota, Penn, Rice, University of Bristol, UCL, University of Houston, and Western Ontario for helpful comments. We are responsible for all errors. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications. © 2016 by Francesco Agostinelli and Matthew Wiswall. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source. Estimating the Technology of Children's Skill Formation Francesco Agostinelli and Matthew Wiswall NBER Working Paper No. 22442 July 2016, Revised March 2022 JEL No. C38,J13 ABSTRACT In this paper we study the process of children’s skill formation. Using a dynamic latent factor structure, we show how measurement restrictions on observed measures aid the identification of skill technology features. We then use our identification results to develop and estimate the joint dynamic process of latent investment and skill development, allowing for static and dynamic complementarities in skill production between parental investments and children’s skills. Using data for the United States, we estimate that parental investments are particularly productive in producing cognitive skills during early childhood (ages 5-6) . Moreover, we find that the marginal productivity of investments in this period is substantially higher for children with lower existing skills, suggesting the optimal targeting of interventions to disadvantaged young children. Francesco Agostinelli University of Pennsylvania 133 South 36th Street Philadelphia, PA 19104 [fagostin@upenn.edu](fagostin@upenn.edu) Matthew Wiswall Department of Economics University of Wisconsin-Madison 1180 Observatory Drive Madison, WI 53706 and NBER [matt.wiswall@gmail.com](matt.wiswall@gmail.com) An online appendix is available at [http://www.nber.org/data-appendix/w22442](http://www.nber.org/data-appendix/w22442) 1 Introduction The wide dispersion of measured human capital in children and its strong correlation with later life outcomes has prompted a renewed interest in understanding the determinants of skill formation among children (for a recent review, see Heckman and Mosso, 2014) . However, the empirical challenges in estimating the skill formation process, principally the technology of child development, is hampered by the imperfect measures of children’s skills we have available. While measurement issues exist in many areas of empirical research, they may be particularly salient in research about child development. There exists a number of different measures of children’s skills, each measure can be arbitrarily located and scaled, and provide widely differing levels of informativeness about the underlying latent skills of the child. In the presence of these measurement issues, identification of the underlying latent process of skill development is particularly challenging, but nonetheless essential, because ignoring the measurement issues through ad ho

FAQs

What problem does the paper address in studying children’s skill formation?

It focuses on estimating the child development technology despite imperfect observed measures of children’s skills, where measurement issues can bias empirical conclusions if ignored.

How does the paper identify features of the skill technology?

It uses a dynamic latent factor structure and imposes empirically grounded measurement restrictions on observed measures to support identification of skill-technology features.

What do the empirical results suggest about parental investments?

Using U.S. data, the paper finds parental investments are especially productive for producing cognitive skills in early childhood (ages 5–6), with higher marginal productivity for children with lower existing skills.

Sep 18, 2026

Uploaded by Grenda