The counterfactual decomposition technique popularized by Blinder (1973, Journal of Human Resources, 436–455) and Oaxaca (1973, International Economic Review, 693–709) is widely used to study mean outcome differences between groups. For example, the technique is often used to analyze wage gaps by sex or race. This article summarizes the technique and addresses several complications, such as the identification of effects of categorical predictors in the detailed decomposition or the estimation of standard errors. A new command called oaxaca is introduced, and examples illustrating its usage are given.

https://journals.sagepub.com/doi/pdf/10.1177/1536867X0800800401

The Blinder–Oaxaca decomposition for linear regression models

This paper empirically analyzes both economic and non-economic determinants of attitudes toward immigrants, within and across countries. The two individual-level survey data sets used, covering a wide range of developed and developing countries, make it possible to test for interactive effects between individual characteristics and country-level attributes. The paper identifies and investigates a strong empirical regularity concerning the relationship between individual skill and attitudes toward immigrants. I find that individuals with higher levels of skill are more likely to be pro-immigration in high per capita GDP countries and less likely in low per capita GDP countries. Additional results, based on a smaller sample of countries, suggest a labor-market explanation for this cross-country pattern. The variation across countries in the correlation between skill and preferences appears to be related to differences in the skill composition of natives relative to immigrants across destination economies. This finding is consistent with the predictions of the Heckscher-Ohlin model, in the absence of factor-price-insensitivity, and of the factor-proportions-analysis model. Finally, non-economic variables also appear to be correlated with immigration attitudes but they do not seem to alter significantly the results on the economic explanations.

Who is Against Immigration? A Cross-Country Investigation of Individual Attitudes Toward Immigrants

Using data from the May and Outgoing Rotation Group (ORG) supplements of the CPS, this paper shows that a large fraction of the growth in residual wage inequality between 1973 and 2003 is due to spurious composition effects. These composition effects are linked to the secular increase in the level of experience and education of the workforce, two factors associated with higher within-group wage dispersion. I also show that both the level and growth in residual wage inequality are overstated in March CPS data that have been used in most previous studies. Measured wages are noisier in the March than in the May/ORG CPS because the March CPS does not measure directly the hourly wages of workers paid by the hour. The extent of measurement error in CPS wages also increases over time. Once these factors are corrected for, I find that residual wage inequality only accounts for a small share of the overall growth in wage inequality. Furthermore, all of the growth in residual wage inequality occurs during the 1980s. This closely mirrors the pattern of change in “between-group” wage differentials like the college-high school wage premium. Overall, the magnitude and timing of the growth in residual wage inequality provides little evidence of a pervasive increase in the demand for skill due, for instance, to skill-biased technological change.

https://www.economics.ubc.ca/files/2013/05/pdf_paper_thomas-lemieux-increasing-residual-wage-inequality.pdf

Increasing Residual Wage Inequality: Composition Effects, Noisy Data, or Rising Demand for Skill?

This paper empirically analyzes economic and noneconomic determinants of individual attitudes toward immigrants, within and across countries. The two survey data sets used, covering a wide range of developed and developing countries, make it possible to test for interactive effects between individual characteristics and country-level attributes. In particular, theory predicts that the correlation between pro-immigration attitudes and individual skill should be related to the skill composition of natives relative to immigrants in the destination country. Skilled individuals should favor immigration in countries where natives are more skilled than immigrants and oppose it otherwise. Results based on direct and indirect measures of the relative skill composition are consistent with these predictions. Noneconomic variables also are correlated with immigration attitudes, but they don't alter significantly the labor-market results.

/pdf/who-is-against-immigration-a-cross-country-investigation-of-ecl419ajk0.pdf

Who Is Against Immigration? A Cross-Country Investigation of Individual Attitudes toward Immigrants

Past research has emphasized two critical economic concerns that appear to generate anti-immigrant sentiment among native citizens: concerns about labor market competition and concerns about the fiscal burden on public services. We provide direct tests of both models of attitude formation using an original survey experiment embedded in a nationwide U.S. survey. The labor market competition model predicts that natives will be most opposed to immigrants who have skill levels similar to their own. We find instead that both low-skilled and highly skilled natives strongly prefer highly skilled immigrants over low-skilled immigrants, and this preference is not decreasing in natives' skill levels. The fiscal burden model anticipates that rich natives oppose low-skilled immigration more than poor natives, and that this gap is larger in states with greater fiscal exposure (in terms of immigrant access to public services). We find instead that rich and poor natives are equally opposed to low-skilled immigration in general. In states with high fiscal exposure, poor (rich) natives are more (less) opposed to low-skilled immigration than they are elsewhere. This indicates that concerns among poor natives about constraints on welfare benefits as a result of immigration are more relevant than concerns among the rich about increased taxes. Overall the results suggest that economic self-interest, at least as currently theorized, does not explain voter attitudes toward immigration. The results are consistent with alternative arguments emphasizing noneconomic concerns associated with ethnocentrism or sociotropic considerations about how the local economy as a whole may be affected by immigration.

/pdf/attitudes-toward-highly-skilled-and-low-skilled-immigration-3j81anptuf.pdf

Attitudes toward Highly Skilled and Low-skilled Immigration: Evidence from a Survey Experiment

We propose a simple methodology for decomposing differences in the first moment into characteristics and coefficients effects. This methodology provides a way to apply the Blinder-Oaxaca decomposition to a non-linear function for both aggregate and detailed decompositions.

https://papers.ssrn.com/sol3/Delivery.cfm?abstractid=446401

Decomposing Differences in the First Moment

Oaxaca and Ransom (1999) show that a detailed decomposition of the coefficients effect is destined to suffer from an identification problem since the detailed coefficients effect attributed to dummy variables is not invariant to the choice of reference groups. It turns out that the identification problem in the decomposition equation is a disguised identification problem of constant and dummy variables in a regression equation. This article proposes a simple and natural remedy for this problem by using "normalized" regressions, which enable me to identify the constant and estimates of each dummy variable.

https://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID435000_code030909590.pdf?abstractid=435000

A Simple Solution to the Identification Problem in Detailed Wage Decompositions

https://www.econstor.eu/bitstream/10419/20112/1/dp877.pdf

Decomposing differences in the first moment

I. INTRODUCTION Oaxaca and Ransom (1999) show that the Oaxaca decomposition suffers from one nagging conceptual problem: The detailed Oaxaca decomposition of wage differentials is not invariant to the choice of reference group when dummy variables are used. (1) That is, if one uses dummy variable(s), then the detailed coefficients effect attributed to dummy variables is not invariant to the choice of the omitted group(s). (2) This invariance or identification problem is well-known to labor economists and has plagued decomposition and discrimination analysis for a long time. (3) This article proposes a simple and practical solution to this identification/invariance problem. The solution is based on the intuitive idea that if alternative reference groups yield different estimates of the characteristics and coefficients effects for each individual variable, then it is natural to obtain estimates of the two effects for every possible specification of the reference groups and take the average of the estimates of the two effects with various reference groups as the "true" contributions of individual variables to wage differentials. A close inspection shows that averaging the two effects in the decomposition equation with varying reference groups (the "averaging approach") is equivalent to finding average estimates of constant and dummy variables in wage equations with varying reference groups and then using the average estimates to calculate the Oaxaca decomposition equation. Researchers may react a bit negatively to this suggestion because it could be quite cumbersome and tedious to estimate wage equations permuting the choice of reference groups. This article shows that there is no need to estimate multiple wage equations. One set of regression estimates is sufficient to resolve the identification problem in the detailed decompositions. That is, the average of the estimates with varying reference groups can be easily calculated by using only "one" set of regression estimates with any specific reference group(s). The identification problem for the detailed decomposition disappears once the contribution of the dummy variables and the constant in the regression equations are identified. (4) II. IDENTIFICATION PROBLEM AND INVARIANT DECOMPOSITION: AN ILLUSTRATION To understand what the identification problem in the detailed wage decomposition is, I will first illustrate the problem with a simple example. Suppose that there are three categories, and the regression equation has only a constant and two dummy variables on the right-hand side where the reference group is the first category. (5) That is, (1) [y.sub.i] = [[alpha].sub.i] + [3.summation over (k=2)] [D.sub.ki] [[beta].sub.ki] + [e.sub.i], where y is log-wages and i = A or B for two comparison groups. Coefficients for two dummy variables taking category one as the reference group are estimated using ordinary least squares (OLS) regression and are reported in the second column of Table 1 for sample A and for sample B. Note that the coefficient of the reference group (category 1) is restricted to zero. Similarly, I may change my reference group specification and estimate the wage equation two more times. The estimates taking category 2 or 3 as the reference group are reported on the next two columns. Because I have three sets of estimates with varying reference groups, I may decompose the wage differentials (0.15 = 0.76 - 0.61) into characteristics and coefficients effects three times. The characteristics effect ([[DELTA].sub.X(1)]) and coefficients effects ([[DELTA].sub.[beta](1)] are specified as follows when category 1 is the reference group: [[DELTA].sub.X(1)] = [[summation].sup.3.sub.k=2] ([[bar.D].sub.kA] - [[bar.D].sub.kB]) [[beta].sub.kB], and [[DELTA].sub.[beta](1)] = [[alpha].sub.A] - [[alpha].sub.B] + [[summation].sup.3.sub.k=2] [[bar.D].sub.kA] ([[beta].sub.kA] - [[beta].sub.kB]). …

/pdf/a-simple-solution-to-the-identification-problem-in-detailed-3s280jd9f9.pdf

A simple solution to the identification problem in detailed wage decompositions

We examine the labor market performance of return migrants using the Hungarian Household Panel Survey. Two distinct selection issues are considered in the estimation of the earnings equation; we implement a natural method using MLE. The result that there is a “premium” to work experience abroad for women is robust across the models we considered. For men, the return to working abroad is not generally significant.

Myeong-Su Yun

Papers

Decomposing Differences in the First Moment

A Simple Solution to the Identification Problem in Detailed Wage Decompositions

Decomposing differences in the first moment

A simple solution to the identification problem in detailed wage decompositions

Returns to returning