1932
0
  1. Home
  2. A-Z Publications
  3. Annual Review of Economics
  4. Volume 10, 2018
  5. Article

Annual Review of Economics

Volume 10, 2018

Identification and Extrapolation of Causal Effects with Instrumental Variables

Abstract

Instrumental variables (IV) are widely used in economics to address selection on unobservables. Standard IV methods produce estimates of causal effects that are specific to individuals whose behavior can be manipulated by the instrument at hand. In many cases, these individuals are not the same as those who would be induced to treatment by an intervention or policy of interest to the researcher. The average causal effect for the two groups can differ significantly if the effect of the treatment varies systematically with unobserved factors that are correlated with treatment choice. We review the implications of this type of unobserved heterogeneity for the interpretation of standard IV methods and for their relevance to policy evaluation. We argue that making inferences about policy-relevant parameters typically requires extrapolating from the individuals affected by the instrument to the individuals who would be induced to treatment by the policy under consideration. We discuss a variety of alternatives to standard IV methods that can be used to rigorously perform this extrapolation. We show that many of these approaches can be nested as special cases of a general framework that embraces the possibility of partial identification.

Keyword(s): causal inference, extrapolation, instrumental variables, JEL C21, JEL C26, JEL C36, JEL C61, local average treatment effect, marginal treatment effects, partial identification
Loading

Article metrics loading...

/content/journals/10.1146/annurev-economics-101617-041813
2018-08-02
2024-04-19
Loading full text...

Full text loading...

/deliver/fulltext/economics/10/1/annurev-economics-101617-041813.html?itemId=/content/journals/10.1146/annurev-economics-101617-041813&mimeType=html&fmt=ahah

Literature Cited

  1. Abadie A, Angrist J, Imbens G 2002. Instrumental variables estimates of the effect of subsidized training on the quantiles of trainee earnings. Econometrica 70:91–117
     [Google Scholar]
  2. Angrist JD 2004. Treatment effect heterogeneity in theory and practice. Econ. J. 114:C52–83
     [Google Scholar]
  3. Angrist JD 2016. Sometimes you get what you need: discussion of Mogstad, Santos, and Torgovitsky Slides presented at the NBER Labor Studies Meeting Cambridge, MA: July 25–29
  4. Angrist JD, Evans WN 1998. Children and their parents' labor supply: evidence from exogenous variation in family size. Am. Econ. Rev. 88:450–77
     [Google Scholar]
  5. Angrist JD, Fernández-Val I 2013. ExtrapoLATE-ing: external validity and overidentification in the LATE framework. Advances in Economics and Econometrics D Acemoglu, M Arellano, E Dekel 401–34 Cambridge, UK: Cambridge Univ. Press
     [Google Scholar]
  6. Angrist JD, Graddy K, Imbens GW 2000. The interpretation of instrumental variables estimators in simultaneous equations models with an application to the demand for fish. Rev. Econ. Stud. 67:499–527
     [Google Scholar]
  7. Angrist JD, Imbens GW 1995. Two-stage least squares estimation of average causal effects in models with variable treatment intensity. J. Am. Stat. Assoc. 90:431–42
     [Google Scholar]
  8. Angrist JD, Krueger AB 1991. Does compulsory school attendance affect schooling and earnings. Q. J. Econ. 106:979–1014
     [Google Scholar]
  9. Angrist JD, Rokkanen M 2015. Wanna get away? Regression discontinuity estimation of exam school effects away from the cutoff. J. Am. Stat. Assoc. 110:1331–44
     [Google Scholar]
  10. Balke A, Pearl J 1997. Bounds on treatment effects from studies with imperfect compliance. J. Am. Stat. Assoc. 92:1171–76
     [Google Scholar]
  11. Bhattacharya J, Shaikh AM, Vytlacil E 2012. Treatment effect bounds: an application to Swan-Ganz catheterization. J. Econom. 168:223–43
     [Google Scholar]
  12. Bitler MP, Gelbach JB, Hoynes HW 2006. What mean impacts miss: distributional effects of welfare reform experiments. Am. Econ. Rev. 96:988–1012
     [Google Scholar]
  13. Bitler MP, Hoynes HW, Domina T 2014. Experimental evidence on distributional effects of Head Start NBER Work. Pap 20434
  14. Björklund A, Moffitt R 1987. The estimation of wage gains and welfare gains in self-selection models. Rev. Econ. Stat. 69:42–49
     [Google Scholar]
  15. Black SE, Devereux PJ, Salvanes KG 2005. The more the merrier? The effect of family size and birth order on children's education. Q. J. Econ. 120:669–700
     [Google Scholar]
  16. Brinch CN, Mogstad M, Wiswall M 2012. Beyond LATE with a discrete instrument Discuss. Pap. 703 Stat. Norway Oslo:
  17. Brinch CN, Mogstad M, Wiswall M 2017. Beyond LATE with a discrete instrument. J. Political Econ. 125:985–1039
     [Google Scholar]
  18. Buchinsky M 1994. Changes in the US wage structure 1963–1987: application of quantile regression. Econometrica 62:405–58
     [Google Scholar]
  19. Carneiro P, Heckman JJ, Vytlacil E 2010. Evaluating marginal policy changes and the average effect of treatment for individuals at the margin. Econometrica 78:377–94
     [Google Scholar]
  20. Carneiro P, Heckman JJ, Vytlacil EJ 2011. Estimating marginal returns to education. Am. Econ. Rev. 101:2754–81
     [Google Scholar]
  21. Carneiro P, Lee S 2009. Estimating distributions of potential outcomes using local instrumental variables with an application to changes in college enrollment and wage inequality. J. Econom. 149:191–208
     [Google Scholar]
  22. Carneiro P, Lokshin M, Umapathi N 2016. Average and marginal returns to upper secondary schooling in Indonesia. J. Appl. Econom. 32:16–36
     [Google Scholar]
  23. Chen X 2007. Large sample sieve estimation of semi-nonparametric models. Handbook of Econometrics 6 JJ Heckman, EE Leamer 5549–632 Amsterdam: Elsevier
     [Google Scholar]
  24. Chernozhukov V, Hansen C 2005. An IV model of quantile treatment effects. Econometrica 73:245–61
     [Google Scholar]
  25. Chernozhukov V, Newey WK, Santos A 2015. Constrained conditional moment restriction models. arXiv: 1509.06311 [math.ST]
  26. Chesher A 2003. Identification in nonseparable models. Econometrica 71:1405–41
     [Google Scholar]
  27. Chesher A 2010. Instrumental variable models for discrete outcomes. Econometrica 78:575–601
     [Google Scholar]
  28. Cornelissen T, Dustmann C, Raute A, Schönberg U 2018. Who benefits from universal child care? Estimating marginal returns to early childcare attendance. J. Political Econ. In press
  29. Dong Y, Lewbel A 2015. Identifying the effect of changing the policy threshold in regression discontinuity models. Rev. Econ. Stat. 97:1081–92
     [Google Scholar]
  30. Doyle JJ Jr. 2007. Child protection and child outcomes: measuring the effects of foster care. Am. Econ. Rev. 97:1583–610
     [Google Scholar]
  31. Dupas P 2014. Short-run subsidies and long-run adoption of new health products: evidence from a field experiment. Econometrica 82:197–228
     [Google Scholar]
  32. Dupas P, Hoffmann V, Kremer M, Zwane AP 2016. Targeting health subsidies through a non-price mechanism: a randomized controlled trial in Kenya. Science 353:889–95
     [Google Scholar]
  33. Eisenhauer P, Heckman JJ, Vytlacil E 2015. The generalized Roy model and the cost-benefit analysis of social programs. J. Political Econ. 123:413–43
     [Google Scholar]
  34. Evans W, Ringel JS 1999. Can higher cigarette taxes improve birth outcomes. J. Public Econ. 72:135–54
     [Google Scholar]
  35. Felfe C, Lalive R 2014. Does early child care help or hurt children's development? Tech. Rep. 8484 Inst. Labor Econ. Bonn, Ger.:
  36. Firpo S, Fortin NM, Lemieux T 2009. Unconditional quantile regressions. Econometrica 77:953–73
     [Google Scholar]
  37. Florens JP, Heckman JJ, Meghir C, Vytlacil E 2008. Identification of treatment effects using control functions in models with continuous, endogenous treatment and heterogeneous effects. Econometrica 76:1191–206
     [Google Scholar]
  38. French E, Song J 2014. The effect of disability insurance receipt on labor supply. Am. Econ. J. Econ. Policy 6:291–337
     [Google Scholar]
  39. Gronau R 1974. Wage comparisons: a selectivity bias. J. Political Econ. 82:1119–43
     [Google Scholar]
  40. Havnes T, Mogstad M 2015. Is universal child care leveling the playing field. J. Public Econ. 127:100–14
     [Google Scholar]
  41. Heckman JJ 1974. Shadow prices, market wages, and labor supply. Econometrica 42:679–94
     [Google Scholar]
  42. Heckman JJ 1976. The common structure of statistical models of truncation, sample selection and limited dependent variables and a simple estimator for such models. Ann. Econ. Soc. Meas. 5:475–92
     [Google Scholar]
  43. Heckman JJ 1979. Sample selection bias as a specification error. Econometrica 47:153–61
     [Google Scholar]
  44. Heckman JJ 1996. Comment. J. Am. Stat. Assoc. 91:459–62
     [Google Scholar]
  45. Heckman JJ 1997. Instrumental variables: a study of implicit behavioral assumptions used in making program evaluations. J. Hum. Resour. 32:441–62
     [Google Scholar]
  46. Heckman JJ 2001. Micro data, heterogeneity, and the evaluation of public policy: nobel lecture. J. Political Econ. 109:673–748
     [Google Scholar]
  47. Heckman JJ 2010. Building bridges between structural and program evaluation approaches to evaluating policy. J. Econ. Lit. 48:356–98
     [Google Scholar]
  48. Heckman JJ, Pinto R 2016. Unordered monotonicity NBER Work. Pap 23497
  49. Heckman JJ, Robb R 1985.a Alternative methods for evaluating the impact of interventions. Longitudinal Analysis of Labor Market Data JJ Heckman, B Singer 156–245 Cambridge, UK: Cambridge Univ. Press
     [Google Scholar]
  50. Heckman JJ, Robb R 1985.b Alternative methods for evaluating the impact of interventions: an overview. J. Econom. 30:239–67
     [Google Scholar]
  51. Heckman JJ, Smith JA 1998. Evaluating the welfare state NBER Work. Pap 6542
  52. Heckman JJ, Smith JA, Clements N 1997. Making the most out of programme evaluations and social experiments: accounting for heterogeneity in programme impacts. Rev. Econ. Stud. 64:487–535
     [Google Scholar]
  53. Heckman JJ, Tobias JL, Vytlacil E 2001. Four parameters of interest in the evaluation of social programs. South. Econ. J. 68:210–23
     [Google Scholar]
  54. Heckman JJ, Tobias JL, Vytlacil E 2003. Simple estimators for treatment parameters in a latent-variable framework. Rev. Econ. Stat. 85:748–55
     [Google Scholar]
  55. Heckman JJ, Urzua S 2010. Comparing IV with structural models: what simple IV can and cannot identify. J. Econom. 156:27–37
     [Google Scholar]
  56. Heckman JJ, Urzua S, Vytlacil E 2006. Understanding instrumental variables in models with essential heterogeneity. Rev. Econ. Stat. 88:389–432
     [Google Scholar]
  57. Heckman JJ, Vytlacil EJ 1999. Local instrumental variables and latent variable models for identifying and bounding treatment effects. PNAS 96:4730–34
     [Google Scholar]
  58. Heckman JJ, Vytlacil EJ 2001.a Policy-relevant treatment effects. Am. Econ. Rev. 91:107–11
     [Google Scholar]
  59. Heckman JJ, Vytlacil EJ 2001.b Instrumental variables, selection models, and tight bounds on the average treatment effect. Econometric Evaluations of Active Labor Market Policies in Europe M Lechner, F Pfeiffer 1–15 Heidelberg, Ger.: Physica
     [Google Scholar]
  60. Heckman JJ, Vytlacil EJ 2001.c Local instrumental variables. Nonlinear Statistical Modeling: Proceedings of the Thirteenth International Symposium in Economic Theory and Econometrics: Essays in Honor of Takeshi Amemiya C Hsiao, K Morimune, JL Powell 1–46 Cambridge, UK: Cambridge Univ. Press
     [Google Scholar]
  61. Heckman JJ, Vytlacil EJ 2005. Structural equations, treatment effects, and econometric policy evaluation. Econometrica 73:669–738
     [Google Scholar]
  62. Heckman JJ, Vytlacil EJ 2007.a Econometric evaluation of social programs, part I: causal models, structural models and econometric policy evaluation. Handbook of Econometrics 6 JJ Heckman, EE Leamer 4779–874 Amsterdam: Elsevier
     [Google Scholar]
  63. Heckman JJ, Vytlacil EJ 2007.b Econometric evaluation of social programs, part II: using the marginal treatment effect to organize alternative econometric estimators to evaluate social programs, and to forecast their effects in new environments. Handbook of Econometrics 6 JJ Heckman, EE Leamer 4875–5143 Amsterdam: Elsevier
     [Google Scholar]
  64. Hull P 2016. Estimating hospital quality with quasi-experimental data Work. Pap. Dep. Econ., Mass. Inst. Technol. Cambridge, MA:
  65. Imbens GW 2010. Better LATE than nothing: some comments on Deaton (2009) and Heckman and Urzua (2009). J. Econ. Lit. 48:399–423
     [Google Scholar]
  66. Imbens GW 2013. Book review feature: Public Policy in an Uncertain World: by Charles F. Manski (Cambridge, MA: Harvard University Press. pp. 224, $39.95. ISBN: 978-0674066892). Econ. J. 123F401–11
     [Google Scholar]
  67. Imbens GW, Angrist JD 1994. Identification and estimation of local average treatment effects. Econometrica 62:467–75
     [Google Scholar]
  68. Imbens GW, Newey WK 2009. Identification and estimation of triangular simultaneous equations models without additivity. Econometrica 77:1481–512
     [Google Scholar]
  69. Imbens GW, Rubin DB 1997. Estimating outcome distributions for compliers in instrumental variables models. Rev. Econ. Stud. 64:555–74
     [Google Scholar]
  70. Kirkeboen LJ, Leuven E, Mogstad M 2016. Field of study, earnings, and self-selection. Q. J. Econ. 131:1057–111
     [Google Scholar]
  71. Kitagawa T 2009. Identification region of the potential outcome distributions under instrument independence Work. Pap. Cent. Microdata Methods Pract. London:
  72. Kitagawa T 2015. A test for instrument validity. Econometrica 83:2043–63
     [Google Scholar]
  73. Kline P, Walters CR 2016. Evaluating public programs with close substitutes: the case of Head Start. Q. J. Econ. 131:1795–848
     [Google Scholar]
  74. Kline P, Walters CR 2017. Through the looking glass: Heckits, LATE, and numerical equivalence Work. Pap. Dep. Econ., Univ. Calif. Berkeley:
  75. Koenker R 2005. Quantile Regression Cambridge, UK: Cambridge Univ. Press
  76. Kowalski A 2016. Doing more when you're running LATE: applying marginal treatment effect methods to examine treatment effect heterogeneity in experiments NBER Work. Pap 22363
  77. Lee S, Salanié B 2016. Identifying effects of multivalued treatments Work. Pap. Dep. Econ., Columbia Univ. New York:
  78. Machado C, Shaikh AM, Vytlacil EJ 2013. Instrumental variables and the sign of the average treatment effect Work. Pap., Dep. Econ., Univ. Chicago
  79. Maestas N, Mullen KJ, Strand A 2013. Does disability insurance receipt discourage work? Using examiner assignment to estimate causal effects of SSDI receipt. Am. Econ. Rev. 103:1797–829
     [Google Scholar]
  80. Manski CF 1989. Anatomy of the selection problem. J. Hum. Resour. 24:343–60
     [Google Scholar]
  81. Manski CF 1990. Nonparametric bounds on treatment effects. Am. Econ. Rev. 80:319–23
     [Google Scholar]
  82. Manski CF 1994. The selection problem. Advances in Econometrics: Sixth World Congress CA Sims 143–70 Cambridge, UK: Cambridge Univ. Press
     [Google Scholar]
  83. Manski CF 1997. Monotone treatment response. Econometrica 65:1311–34
     [Google Scholar]
  84. Manski CF 2003. Partial Identification of Probability Distributions Berlin: Springer
  85. Manski CF, Pepper JV 2000. Monotone instrumental variables: with an application to the returns to schooling. Econometrica 68:997–1010
     [Google Scholar]
  86. Manski CF, Pepper JV 2009. More on monotone instrumental variables. Econom. J. 12:S200–16
     [Google Scholar]
  87. Masten MA 2015. Random coefficients on endogenous variables in simultaneous equations models Work. Pap. 25/15 Cent. Microdata Methods Pract. London:
  88. Masten MA, Torgovitsky A 2016. Identification of instrumental variable correlated random coefficients models. Rev. Econom. Stat. 98:1001–5
     [Google Scholar]
  89. Matzkin RL 2007. Nonparametric identification. Handbook of Econometrics 6 JJ Heckman, EE Leamer 5307–68 Amsterdam: Elsevier
     [Google Scholar]
  90. Miguel E, Satyanath S, Sergenti E 2004. Economic shocks and civil conflict: an instrumental variables approach. J. Political Econ. 112:725–53
     [Google Scholar]
  91. Moffitt R 2008. Estimating marginal treatment effects in heterogeneous populations. Ann. Econ. Stat. 91/92:239–61
     [Google Scholar]
  92. Mogstad M, Santos A, Torgovitsky A 2017. Using instrumental variables for inference about policy relevant treatment parameters NBER Work. Pap 23568
  93. Mourifié I 2015. Sharp bounds on treatment effects in a binary triangular system. J. Econom. 187:74–81
     [Google Scholar]
  94. Nybom M 2017. The distribution of lifetime earnings returns to college. J. Labor Econ. 35:903–52
     [Google Scholar]
  95. Robinson PM 1988. Root-N-consistent semiparametric regression. Econometrica 56:931–54
     [Google Scholar]
  96. Rokkanen M 2015. Exam schools, ability, and the effects of affirmative action: latent factor extrapolation in the regression discontinuity design Work. Pap., Dep. Econ., Columbia Univ. New York:
  97. Shaikh AM, Vytlacil EJ 2011. Partial identification in triangular systems of equations with binary dependent variables. Econometrica 79:949–55
     [Google Scholar]
  98. Tamer E 2010. Partial identification in econometrics. Annu. Rev. Econ. 2:167–95
     [Google Scholar]
  99. Torgovitsky A 2015. Identification of nonseparable models using instruments with small support. Econometrica 83:1185–97
     [Google Scholar]
  100. Torgovitsky A 2017.a Minimum distance from independence estimation of nonseparable instrumental variables models. J. Econom. 199:35–48
     [Google Scholar]
  101. Torgovitsky A 2017.b Partial identification by extending subdistributions Work. Pap., Univ. Chicago
  102. Vuong Q, Xu H 2017. Counterfactual mapping and individual treatment effects in nonseparable models with binary endogeneity. Quant. Econ. 8:589–610
     [Google Scholar]
  103. Vytlacil E 2002. Independence, monotonicity, and latent index models: an equivalence result. Econometrica 70:331–41
     [Google Scholar]
  104. Vytlacil E, Yildiz N 2007. Dummy endogenous variables in weakly separable models. Econometrica 75:757–79
     [Google Scholar]
  105. Walters C 2014. The demand for effective charter schools NBER Work. Pap 20640
  106. Wing C, Cook TD 2013. Strengthening the regression discontinuity design using additional design elements: a within-study comparison. J. Policy Anal. Manag. 32:853–77
     [Google Scholar]
/content/journals/10.1146/annurev-economics-101617-041813
Loading
Identification and Extrapolation of Causal Effects with Instrumental Variables
Annual Review of Economics 10, 577 (2018); https://doi.org/10.1146/annurev-economics-101617-041813
/content/journals/10.1146/annurev-economics-101617-041813
/content/journals/10.1146/annurev-economics-101617-041813
Loading

Data & Media loading...

  • Article Type: Review Article

Most Cited Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error