nach oben

Empirical Economics

Erschienen in:

22.11.2018

Consistent estimates of the public/private wage gap

verfasst von: Domenico Depalo, Santiago Pereda-Fernández

Erschienen in: Empirical Economics | Ausgabe 6/2020

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

Existing estimates of the public/private wage gap allow for possible sorting of individuals into one sector, but they rely on parametric assumptions that may introduce substantial bias in the parameter of interest. Solutions are semi- and nonparametric approaches. For Italy, the latter methods yield a gap of approximately 20%, whereas the bias from parametric assumptions is as large as 10%.

Vorheriger Artikel Monetary shocks and job flows: evidence from disaggregated data

Nächster Artikel Is there an informal employment wage penalty in Egypt? Evidence from quantile regression on panel data

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Nur mit Berechtigung zugänglich

In this note, we assume the existence of an exclusion restriction. However, this is not strictly required in the fully parametric model.

For illustrative purposes, we provide formal intuition on the size of the bias of the Heckman two-step estimator in “Appendix A.”

Newey (2009) considers power series and splines for the second stage (denoted by \(\left\{ \cdot \right\} _{k=1}^{K}\)), and some semiparametric methods to estimate the first stage. For a more transparent comparison to Das et al. (2003), we use power series in the wage equation and the linear probability model in the selection equation.

For a review on sample selection estimators, see Vella (1998).

The test takes the form of two inequalities that are necessary to identify a Local Average Treatment Effect (Imbens and Angrist 1994): \(P[y,D=1|Z=0] \le P[y,D=1|Z=1]\) and \(P[y,D=0|Z=1] \le P[y,D=0|Z=0]\). If any of the two inequalities is violated, the validity of the instrument is falsified. In the latter case, the test is not informative about which of the two assumptions fails.

As a further check in favor of the validity of the instrument, we estimate the model with IV using a more parsimonious and plausibly exogenous set of controls and then add them progressively: Under the maintained assumption of constant treatment effect, finding stable results would be additional evidence in favor of the validity of the instrument and the method(s) in general; if there is instability in the estimated effect, nothing can be said. Controlling only for age education and year, the wage differential is slightly above 30% (Table 6); as soon as we add the job position, it marginally decreases to 28%, where it remains no matter what variables we add.

Standard errors for the OLS, IV, and Heckman estimators were calculated using the asymptotic formula; those for the Newey (2009) and Das et al. (2003) estimators were bootstrapped.

The difference between the two estimators could be due to the relevance of the exclusion restriction (Puhani 2000). In our data, the F-statistic is equal to 257.6, a very large value (Bound et al. 1995). Therefore, we argue that the distribution assumptions play a major role.

Note that the top and bottom 1% of the sample is trimmed in the estimation of the semi- and nonparametric estimators to guarantee that the propensity score estimates are strictly inside the unit interval.

The conditions are those of the IV for local average treatment effect (Imbens and Angrist 1994), as well as bounded outcome, rank similarity, and first-order stochastic dominance.

The exact meaning of “large” is an individual choice of the researcher that should trade off the benefits from a simpler communication and the costs from unduly restrictive assumptions.

The relevance of the instrument is less important for partial than for point identification, because the set is valid for the entire population rather than for compliers only; see Chen et al. (2017) for a thorough discussion.

Arabmazar A, Schmidt P (1982) An investigation of the robustness of the tobit estimator to non-normality. Econometrica 50(4):1055–1063CrossRef

Bhattacharya J, Shaikh AM, Vytlacil E (2008) Treatment effect bounds under monotonicity assumptions: an application to Swan-Ganz catheterization. Am Econ Rev 98(2):351–56CrossRef

Bhattacharya J, Shaikh AM, Vytlacil E (2012) Treatment effect bounds: an application to SwanGanz catheterization. J Econom 168(2):223–243CrossRef

Bound J, Jaeger D, Baker R (1995) Problems with instrumental variable estimation when the correlation between the instruments and the endogenous explanatory variable is weak. J Am Stat Assoc 90:443–450

Brunello G, Dustmann C (1997) Le retribuzioni nei settori pubblico e privato in Italia e in Germania: un paragone basato su dati microeconomici. In: Dell’Aringa C (ed) Rapporto ARAN sulle retribuzioni. ARAN, Franco Angeli

Card D (1999) The causal effect of education on earnings. In: Ashenfelter O, Card D (eds) Handbook of labor economics, volume 3 of handbook of labor economics. Elsevier, New York, pp 1801–1863CrossRef

Chen X, Flores CA, Flores-Lagunes A (2017) Going beyond LATE: bounding average treatment effects of job corps training. J Hum Resour (forthcoming)

Das M, Newey WK, Vella F (2003) Nonparametric estimation of sample selection models. Rev Econ Stud 70(1):33–58CrossRef

Depalo D (2018) Identification issues in the public/private wage gap with an application to Italy. J Appl Econom 33(3):435–456CrossRef

Depalo D, Giordano R (2011) The public–private pay gap: a robust quantile approach. Giornale degli Economisti 70(1):25–64

Dustmann C, van Soest A (1998) Public and private sector wages of male workers in Germany. Eur Econ Rev 42(8):1417–1441CrossRef

Heckman JJ (1979) Sample selection bias as a specification error. Econom Econom Soc 47(1):153–61CrossRef

Heckman JJ, Vytlacil EJ (2007) Econometric evaluation of social programs, Part I: causal models, structural models and econometric policy evaluation. In: Heckman J, Leamer E (eds) Handbook of econometrics, volume 6 of handbook of econometrics. Elsevier, New York

Imbens GW, Angrist JD (1994) Identification and estimation of local average treatment effects. Econometrica 62(2):467–75CrossRef

Imbens GW, Rubin DB (1997) Estimating outcome distributions for compliers in instrumental variables models. Rev Econ Stud 64(4):555–574CrossRef

Lausěv J (2014) What has 20 years of publicprivate pay gap literature told us? Eastern european transitioning vs. developed economies. J Econ Surv 28(3):516–550CrossRef

Lucifora C, Meurs D (2006) The public sector pay gap in France, Great Britain and Italy. Rev Income Wealth 52(1):43–59CrossRef

Maddala GS (1983) Limited-dependent and qualitative variables in econometrics. Cambridge University Press, CambridgeCrossRef

Mourifié I, Wan Y (2017) Testing local average treatment effect assumptions. Rev Econ Stat 99(2):305–313CrossRef

Newey WK (1999) Consistency of two-step sample selection estimators despite misspecification of distribution. Econ Lett 63(2):129–132CrossRef

Newey WK (2009) Two-step series estimation of sample selection models. Econom J 12:S217–S229CrossRef

Nijman T, Verbeek M (1992) Nonresponse in panel data: the impact on estimates of a life cycle consumption function. J Appl Econom 7(3):243–257CrossRef

Puhani P (2000) The Heckman correction for sample selection and its critique. J Econ Surv 14(1):53–68CrossRef

Vella F (1998) Estimating models with sample selection bias: a survey. J Hum Resour 33(1):127–169CrossRef

Vytlacil E (2002) Independence, monotonicity, and latent index models: an equivalence result. Econometrica 70(1):331–341CrossRef

Titel: Consistent estimates of the public/private wage gap
verfasst von: Domenico Depalo
Santiago Pereda-Fernández
Publikationsdatum: 22.11.2018
Verlag: Springer Berlin Heidelberg
Erschienen in: Empirical Economics / Ausgabe 6/2020
Print ISSN: 0377-7332
Elektronische ISSN: 1435-8921
DOI: https://doi.org/10.1007/s00181-018-1592-7

Springer Professional

Abstract

Bitte loggen Sie sich ein, um Zugang zu Ihrer Lizenz zu erhalten.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft"

Springer Professional "Wirtschaft+Technik"

Weitere Artikel der Ausgabe 6/2020

The politics of environmental enforcement: the case of the Resource and Conservation Recovery Act

Why fully efficient banks matter? A nonparametric stochastic frontier approach in the presence of fully efficient banks

Socioeconomic factors and happiness: evidence from self-reported mental health data

The effects of economic policy uncertainty on European economies: evidence from a TVP-FAVAR

Lost economies of scope and potential merger gains in the Norwegian electricity industry

Income inequality, financial flows and political institution: sub-Saharan African financial network

Premium Partner