{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,13]],"date-time":"2026-09-13T08:03:25Z","timestamp":1789286605535,"version":"build-2803163510"},"reference-count":54,"publisher":"MIT Press","issue":"2","content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2024,3,19]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>This study proposes an econometric framework to interpret and empirically decompose the difference between instrumental variables (IV) and ordinary least squares (OLS) estimates given by a linear regression model when the true causal effects of the treatment are nonlinear in treatment levels and heterogeneous across covariates. I show that the IV-OLS coefficient gap consists of three estimable components: the difference in weights on the covariates, the difference in weights on the treatment levels, and the difference in identified marginal effects that arises from endogeneity bias. Applications of this framework to return-to-schooling estimates demonstrate the empirical relevance of this distinction in properly interpreting the IV-OLS gap.<\/jats:p>","DOI":"10.1162\/rest_a_01169","type":"journal-article","created":{"date-parts":[[2022,2,4]],"date-time":"2022-02-04T00:38:54Z","timestamp":1643935134000},"page":"505-520","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":27,"title":["Empirical Decomposition of the IV-OLS Gap with Heterogeneous and Nonlinear Effects"],"prefix":"10.1162","volume":"106","author":[{"given":"Shoya","family":"Ishimaru","sequence":"first","affiliation":[{"name":"Hitotsubashi University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"281","published-online":{"date-parts":[[2024,3,19]]},"reference":[{"key":"2024032019203919100_bib1","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1086\/654403","article-title":"How Large Are Human-Capital Externalities? Evidence from Compulsory Schooling Laws","volume":"15","author":"Acemoglu","year":"2000","journal-title":"NBER Macroeconomics Annual"},{"issue":"2","key":"2024032019203919100_bib2","first-page":"481","article-title":"A Practical Asymptotic Variance Estimator for Two-Step Semiparametric Estimators","volume":"94","author":"Ackerberg","year":"2012"},{"key":"2024032019203919100_bib3","doi-asserted-by":"publisher","first-page":"249","DOI":"10.2307\/2998558","article-title":"Estimating the Labor Market Impact of Voluntary Military Service Using Social Security Data on Military Applicants","volume":"66","author":"Angrist","year":"1998","journal-title":"Econometrica"},{"issue":"3","key":"2024032019203919100_bib4","doi-asserted-by":"publisher","first-page":"499","DOI":"10.1111\/1467-937X.00141","article-title":"The Interpretation of Instrumental Variables Estimators in Simultaneous Equations Models with an Application to the Demand for Fish","volume":"67","author":"Angrist","year":"2000","journal-title":"The Review of Economic Studies"},{"issue":"430","key":"2024032019203919100_bib5","doi-asserted-by":"publisher","first-page":"431","DOI":"10.1080\/01621459.1995.10476535","article-title":"Two-Stage Least Squares Estimation of Average Causal Effects in Models with Variable Treatment Intensity","volume":"90","author":"Angrist","year":"1995","journal-title":"Journal of the American Statistical Association"},{"issue":"4","key":"2024032019203919100_bib6","doi-asserted-by":"publisher","first-page":"979","DOI":"10.2307\/2937954","article-title":"Does Compulsory School Attendance Affect Schooling and Earnings?","volume":"106","author":"Angrist","year":"1991","journal-title":"The Quarterly Journal of Economics"},{"key":"2024032019203919100_bib7","doi-asserted-by":"publisher","first-page":"1277","DOI":"10.1016\/S1573-4463(99)03004-7","article-title":"Empirical Strategies in Labor Economics","volume-title":"Handbook of Labor Economics","author":"Angrist","year":"1999"},{"issue":"2","key":"2024032019203919100_bib8","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1257\/jep.24.2.3","article-title":"The Credibility Revolution in Empirical Economics: How Better Research Design Is Taking the Con Out of Econometrics","volume":"24","author":"Angrist","year":"2010","journal-title":"Journal of Economic Perspectives"},{"issue":"1","key":"2024032019203919100_bib9","doi-asserted-by":"publisher","first-page":"250","DOI":"10.1111\/ajps.12185","article-title":"Does Regression Produce Representative Estimates of Causal Effects?","volume":"60","author":"Aronow","year":"2016","journal-title":"American Journal of Political Science"},{"key":"2024032019203919100_bib10","volume-title":"Human Capital and the Personal Distribution of Income: An Analytical Approach","author":"Becker","year":"1967"},{"issue":"1","key":"2024032019203919100_bib11","doi-asserted-by":"publisher","first-page":"437","DOI":"10.1257\/0002828053828635","article-title":"Why the Apple Doesn\u2019t Fall Far: Understanding Intergenerational Transmission of Human Capital","volume":"95","author":"Black","year":"2005","journal-title":"American Economic Review"},{"issue":"6","key":"2024032019203919100_bib12","doi-asserted-by":"publisher","first-page":"1613","DOI":"10.1111\/j.1468-0262.2007.00808.x","article-title":"Semi-Nonparametric IV Estimation of Shape-Invariant Engel Curves","volume":"75","author":"Blundell","year":"2007","journal-title":"Econometrica"},{"issue":"4","key":"2024032019203919100_bib13","doi-asserted-by":"publisher","first-page":"1035","DOI":"10.1111\/j.1467-937X.2007.00458.x","article-title":"A Non-Parametric Test of Exogeneity","volume":"74","author":"Blundell","year":"2007","journal-title":"The Review of Economic Studies"},{"issue":"1","key":"2024032019203919100_bib14","doi-asserted-by":"publisher","first-page":"132","DOI":"10.1086\/379937","article-title":"Estimation of Educational Borrowing Constraints Using Returns to Schooling","volume":"112","author":"Cameron","year":"2004","journal-title":"Journal of Political Economy"},{"key":"2024032019203919100_bib15","volume-title":"Earnings, Schooling, and Ability Revisited","author":"Card","year":"1995"},{"key":"2024032019203919100_bib16","volume-title":"Using Geographic Variation in College Proximity to Estimate the Return to Schooling","author":"Card","year":"1995"},{"key":"2024032019203919100_bib17","doi-asserted-by":"publisher","first-page":"1801","DOI":"10.1016\/S1573-4463(99)03011-4","article-title":"The Causal Effect of Education on Earnings","volume-title":"Handbook of Labor Economics","author":"Card","year":"1999"},{"issue":"5","key":"2024032019203919100_bib18","doi-asserted-by":"publisher","first-page":"1127","DOI":"10.1111\/1468-0262.00237","article-title":"Estimating the Return to Schooling: Progress on Some Persistent Econometric Problems","volume":"69","author":"Card","year":"2001","journal-title":"Econometrica"},{"issue":"1","key":"2024032019203919100_bib19","doi-asserted-by":"publisher","first-page":"377","DOI":"10.3982\/ECTA7089","article-title":"Evaluating Marginal Policy Changes and the Average Effect of Treatment for Individuals at the Margin","volume":"78","author":"Carneiro","year":"2010","journal-title":"Econometrica"},{"issue":"6","key":"2024032019203919100_bib20","doi-asserted-by":"publisher","first-page":"2754","DOI":"10.1257\/aer.101.6.2754","article-title":"Estimating Marginal Returns to Education","volume":"101","author":"Carneiro","year":"2011","journal-title":"American Economic Review"},{"issue":"1","key":"2024032019203919100_bib23","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1016\/j.jeconom.2006.06.002","article-title":"Instrumental Variable Estimation of Nonseparable Models","volume":"139","author":"Chernozhukov","year":"2007","journal-title":"Journal of Econometrics"},{"issue":"2","key":"2024032019203919100_bib21","doi-asserted-by":"publisher","first-page":"999","DOI":"10.1093\/restud\/rdx049","article-title":"Fuzzy Differences-in-Differences","volume":"85","author":"de Chaisemartin","year":"2018","journal-title":"The Review of Economic Studies"},{"issue":"9","key":"2024032019203919100_bib22","doi-asserted-by":"publisher","first-page":"2964","DOI":"10.1257\/aer.20181169","article-title":"Two-Way Fixed Effects Estimators with Heterogeneous Treatment Effects","volume":"110","author":"de Chaisemartin","year":"2020","journal-title":"American Economic Review"},{"issue":"3","key":"2024032019203919100_bib24","doi-asserted-by":"crossref","first-page":"395","DOI":"10.1080\/07350015.2014.888290","article-title":"Testing the Unconfoundedness Assumption via Inverse Probability Weighted Estimators of (L)ATT","volume":"32","author":"Donald","year":"2014","journal-title":"Journal of Business & Economic Statistics"},{"issue":"4","key":"2024032019203919100_bib25","doi-asserted-by":"publisher","first-page":"795","DOI":"10.1257\/aer.91.4.795","article-title":"Schooling and Labor Market Consequences of School Construction in Indonesia: Evidence from an Unusual Policy Experiment","volume":"91","author":"Duflo","year":"2001","journal-title":"American Economic Review"},{"key":"2024032019203919100_bib26","doi-asserted-by":"publisher","first-page":"23","DOI":"10.2307\/1401917","article-title":"Errors in Variables","volume":"22","author":"Durbin","year":"1954","journal-title":"Review of the International Statistical Institute"},{"issue":"3","key":"2024032019203919100_bib27","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1080\/07350015.2017.1366909","article-title":"Why High-Order Polynomials Should Not Be Used in Regression Discontinuity Designs","volume":"37","author":"Gelman","year":"2019","journal-title":"Journal of Business & Economic Statistics"},{"key":"2024032019203919100_bib28","doi-asserted-by":"publisher","first-page":"1251","DOI":"10.2307\/1913827","article-title":"Specification Tests in Econometrics","volume":"46","author":"Hausman","year":"1978","journal-title":"Econometrica"},{"issue":"1","key":"2024032019203919100_bib29","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1016\/j.jeconom.2009.09.006","article-title":"Comparing IV with Structural Models: What Simple IV Can and Cannot Identify","volume":"156","author":"Heckman","year":"2010","journal-title":"Journal of Econometrics"},{"issue":"3","key":"2024032019203919100_bib30","first-page":"389","article-title":"Understanding Instrumental Variables in Models with Essential Heterogeneity","volume":"88","author":"Heckman","year":"2006"},{"issue":"2","key":"2024032019203919100_bib31","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1257\/aer.91.2.107","article-title":"Policy-Relevant Treatment Effects","volume":"91","author":"Heckman","year":"2001","journal-title":"American Economic Review"},{"issue":"2","key":"2024032019203919100_bib32","doi-asserted-by":"publisher","first-page":"347","DOI":"10.3982\/ECTA8662","article-title":"Applied Nonparametric Instrumental Variables Estimation","volume":"79","author":"Horowitz","year":"2011","journal-title":"Econometrica"},{"issue":"2","key":"2024032019203919100_bib33","doi-asserted-by":"publisher","first-page":"467","DOI":"10.2307\/2951620","article-title":"Identification and Estimation of Local Average Treatment Effects","volume":"62","author":"Imbens","year":"1994","journal-title":"Econometrica"},{"issue":"5","key":"2024032019203919100_bib34","doi-asserted-by":"publisher","first-page":"1481","DOI":"10.3982\/ECTA7108","article-title":"Identification and Estimation of Triangular Simultaneous Equations Models without Additivity","volume":"77","author":"Imbens","year":"2009","journal-title":"Econometrica"},{"issue":"3","key":"2024032019203919100_bib35","first-page":"600","article-title":"Labor-Market Returns to Two-and Four-Year College","volume":"85","author":"Kane","year":"1995","journal-title":"The American Economic Review"},{"issue":"3","key":"2024032019203919100_bib36","doi-asserted-by":"crossref","first-page":"358","DOI":"10.1198\/073500101681020006","article-title":"Interpreting Instrumental Variables Estimates of the Returns to Schooling","volume":"19","author":"Kling","year":"2001","journal-title":"Journal of Business & Economic Statistics"},{"key":"2024032019203919100_bib37","article-title":"Ability Bias, Discount Rate Bias and the Return to Education","author":"Lang","year":"1993"},{"key":"2024032019203919100_bib38","doi-asserted-by":"crossref","DOI":"10.3386\/w8605","article-title":"The Effect of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports","author":"Lochner","year":"2001"},{"issue":"2","key":"2024032019203919100_bib39","first-page":"387","article-title":"Estimating and Testing Models with Many Treatment Levels and Limited Instruments","volume":"97","author":"Lochner","year":"2015"},{"issue":"2","key":"2024032019203919100_bib40","first-page":"1","article-title":"What Linear Estimators Miss: The Effects of Family Income on Child Outcomes","volume":"4","author":"L\u00f8ken","year":"2012","journal-title":"American Economic Journal: Applied Economics"},{"issue":"5","key":"2024032019203919100_bib41","first-page":"1001","article-title":"Identification of Instrumental Variable Correlated Random Coefficients Models","volume":"98","author":"Masten","year":"2016"},{"issue":"5","key":"2024032019203919100_bib42","doi-asserted-by":"publisher","first-page":"1589","DOI":"10.3982\/ECTA15463","article-title":"Using instrumental variables for inference about policy relevant treatment parameters","volume":"86","author":"Mogstad","year":"2018","journal-title":"Econometrica"},{"issue":"11","key":"2024032019203919100_bib43","doi-asserted-by":"publisher","first-page":"3663","DOI":"10.1257\/aer.20190221","article-title":"The Causal Interpretation of Two-Stage Least Squares with Multiple Instrumental Variables","volume":"111","author":"Mogstad","year":"2021","journal-title":"American Economic Review"},{"key":"2024032019203919100_bib44","doi-asserted-by":"crossref","DOI":"10.2139\/ssrn.1686527","article-title":"Linearity in Instrumental Variables Estimation: Problems and Solutions","author":"Mogstad","year":"2010"},{"key":"2024032019203919100_bib45","doi-asserted-by":"publisher","first-page":"2111","DOI":"10.1016\/S1573-4412(05)80005-4","article-title":"Large Sample Estimation and Hypothesis Testing","volume-title":"Handbook of Econometrics","author":"Newey","year":"1994"},{"issue":"5","key":"2024032019203919100_bib46","doi-asserted-by":"publisher","first-page":"1565","DOI":"10.1111\/1468-0262.00459","article-title":"Instrumental Variable Estimation of Nonparametric Models","volume":"71","author":"Newey","year":"2003","journal-title":"Econometrica"},{"issue":"1","key":"2024032019203919100_bib47","doi-asserted-by":"publisher","first-page":"152","DOI":"10.1257\/000282806776157641","article-title":"Estimating Average and Local Average Treatment Effects of Education When Compulsory Schooling Laws Really Matter","volume":"96","author":"Oreopoulos","year":"2006","journal-title":"American Economic Review"},{"key":"2024032019203919100_bib48","article-title":"Estimating Average and Local Average Treatment Effects of Education When Compulsory Schooling Laws Really Matter: Corrigendum","author":"Oreopoulos","year":"2008","journal-title":"American Economic Review"},{"key":"2024032019203919100_bib49","first-page":"287","article-title":"The Gini Instrumental Variable, or the \u201cDouble Instrumental Variable\u201d Estimator","volume":"62","author":"Schechtman","year":"2004","journal-title":"Metron"},{"key":"2024032019203919100_bib50","doi-asserted-by":"crossref","DOI":"10.2139\/ssrn.3846693","article-title":"When Should We (Not) Interpret Linear IV Estimands as LATE?","author":"S\u0142oczy\u0144ski","year":"2021"},{"issue":"3","key":"2024032019203919100_bib51","first-page":"501","article-title":"Interpreting OLS Estimands When Treatment Effects Are Heterogeneous: Smaller Groups Get Larger Weights","volume":"104","author":"S\u0142oczy\u0144ski","year":"2022"},{"issue":"3","key":"2024032019203919100_bib52","doi-asserted-by":"publisher","first-page":"557","DOI":"10.2307\/2171753","article-title":"Instrumental Variables Regression with Weak Instruments","volume":"65","author":"Staiger","year":"1997","journal-title":"Econometrica"},{"issue":"4","key":"2024032019203919100_bib53","doi-asserted-by":"publisher","first-page":"733","DOI":"10.2307\/1914093","article-title":"Alternative Tests of Independence between Stochastic Regressors and Disturbances","volume":"41","author":"Wu","year":"1973","journal-title":"Econometrica"},{"issue":"4","key":"2024032019203919100_bib54","doi-asserted-by":"crossref","first-page":"478","DOI":"10.1080\/07350015.1996.10524677","article-title":"On Using Linear Regressions in Welfare Economics","volume":"14","author":"Yitzhaki","year":"1996","journal-title":"Journal of Business & Economic Statistics"}],"container-title":["Review of Economics and Statistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/rest\/article-pdf\/106\/2\/505\/2351603\/rest_a_01169.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/rest\/article-pdf\/106\/2\/505\/2351603\/rest_a_01169.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,20]],"date-time":"2024-03-20T19:20:56Z","timestamp":1710962456000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/rest\/article\/106\/2\/505\/109261\/Empirical-Decomposition-of-the-IV-OLS-Gap-with"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,3]]},"references-count":54,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2024,3,19]]},"published-print":{"date-parts":[[2024,3,19]]}},"URL":"https:\/\/doi.org\/10.1162\/rest_a_01169","relation":{},"ISSN":["0034-6535","1530-9142"],"issn-type":[{"value":"0034-6535","type":"print"},{"value":"1530-9142","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2024,3]]},"published":{"date-parts":[[2024,3]]}}}