Abstract:

In recent decades, the number of women pursuing careers in health has significantly increased. However, the physician labor market is still characterized by gender differences regarding payment. Using a nationally representative Peruvian sample of health providers (3,219 male and 1,063 female physicians), we estimated the gender gap in the likelihood of earning high wages for physicians and decomposed this gap in a proportion related to differences in individual characteristics (e.g. specialty, labor experience), and a residual proportion related to differences in returns to these characteristics. Our main results reveal that male physicians have on average an 81% higher likelihood of earning high salaries (monthly earning level > 5,000 PEN) relative to their female counterparts. Further, the main proportion of this gap is associated to the unexplained component (among 57% and 77%, according to the model specification), which may be associated to unobservable characteristics and discrimination in the Peruvian labor market.

Keywords:
Physicians; Wages; Gender Gap

Resumen:

En décadas recientes, el número de mujeres realizando su carrera en el ámbito de salud se ha incrementado significativamente. No obstante, el mercado laboral de los médicos está todavía caracterizado por diferencias de género respecto a los salarios. Utilizando una muestra peruana nacionalmente representativa de proveedores de salud (3.219 hombres y 1.063 mujeres médicos), estimamos la brecha de género en la probabilidad para los médicos de ganar sueldos altos y la desglosamos según los porcentajes vinculados a las diferencias relacionadas con las características individuales (p.ej. especialidad, experiencia laboral) y un porcentaje residual vinculado a las diferencias relacionadas con estas características. Nuestros resultados principales revelaron que los médicos hombres contaban en promedio con un 81% mayor probabilidad de ganar sueldos más altos (nivel mensual de renta > 5.000 PEN) frente a sus compañeras mujeres. Asimismo, gran parte del porcentaje de esta brecha está asociado a un componente inexplicable (entre un 57% y un 77%, según la especificación del modelo), lo que tal vez esté relacionado con las características no observables y la discriminación en el mercado laboral peruano.

Palabras-clave:
Médicos; Sueldos; Sesgo de Género

Resumo:

Nas últimas décadas, o número de mulheres atuando em carreiras da saúde aumentou significativamente. Contudo, o mercado de trabalho médico continua caracterizado por diferenças de gênero nos salários. Usando uma amostra nacional representativa de profissionais da saúde peruanos (3.219 médicos e 1.063 médicas), nós estimamos a diferença de gênero na probabilidade de receber altos salários para médicos e decompomos essa diferença em uma proporção relacionada a diferenças em características individuais (p.ex.: especialidade, experiência profissional) e uma proporção residual relacionada a diferenças de retornos dessas características. Nossos resultados principais revelam que os médicos têm, em média, uma probabilidade 81% maior de receber salários altos (nível de rendimentos mensais > 5.000 PEN) em relação às médicas. Adicionalmente, a principal proporção dessa diferença está associada ao componente não-explicado (entre 57% e 77%, de acordo com a especificação do modelo), o que pode estar associado a características não-observadas e discriminação no mercado de trabalho peruano.

Palavras-chave:
Médicos; Soldos; Lacuna de Gênero

Introduction

In recent decades, the number of women pursuing careers in health has significantly increased ¹1. Jolliff L, Leadley J, Coakley E, Sloane RA. Women in US academic medicine and science: statistics and benchmarking report 2011-2012. Washington DC: Association of American Medical Colleges; 2012.. However, health occupations are still characterized by significant gender differences. Particularly in the physician labor market, differences in payment by gender persist as a main concern. For instance, during the period 2006-2010 in the United States, male physicians earned on average 25.3% more than their female counterparts ²2. Seabury SA, Chandra A, Jena AB. Trends in the earnings of male and female health care professionals in the United States, 1987 to 2010. JAMA Intern Med 2013; 173:1748-50.. Furthermore, the gender wage gap trends are maintained for specific public health systems ³3. Theurl E, Winner H. The male-female gap in physician earnings: evidence from a public health insurance system. Health Econ 2011; 20:1184-200. and specific specialties ⁴4. Jagsi R, Griffith KA, Stewart A, Sambuco D, DeCastro R, Ubel PA. Gender differences in the salaries of physician researchers. JAMA 2012; 307:2410-7..

Many reasons explain why women in medicine are paid less. First, they are mainly concentrated in lower paying specialties, such as pediatrics ⁵5. Weeks WB, Wallace TA, Wallace AE. How do race and sex affect the earnings of primary care physicians? Health Aff (Millwood) 2009; 28:557-66., while men are concentrated in higher paying specialties, such as surgical ⁴4. Jagsi R, Griffith KA, Stewart A, Sambuco D, DeCastro R, Ubel PA. Gender differences in the salaries of physician researchers. JAMA 2012; 307:2410-7.^,66. Singh A, Burke CA, Larive B, Sastri SV. Women in Gastroenterology Committee of American College of Gastroenterology. Do gender disparities persist in gastroenterology after 10 years of practice? Am J Gastroenterol 2008; 103:1589-95.. Second, there are differences in productivity among male and female physicians. On average, women in the health field worked fewer hours, therefore they attend fewer patients than their male counterparts ³3. Theurl E, Winner H. The male-female gap in physician earnings: evidence from a public health insurance system. Health Econ 2011; 20:1184-200.. Finally, the prejudices and discrimination against female physicians might be a potential explanation. There is still a huge gap in financial compensation between male and female physicians, even when they do the same activity and work the same hours ⁷7. Bashaw DJ, Heywood JS. The gender earnings gap for US physicians: has equality been achieved? Labour 2001; 15:371-91.^,88. Lo Sasso AT, Richards MR, Chou C-F, Gerber SE. The $16,819 pay gap for newly trained physicians: the unexplained trend of men earning more than women. Health Aff (Millwood) 2011; 30:193-201.^,99. Esteves-Sorenson C, Snyder J. The gender earnings gap for physicians and its increase over time. Econ Lett 2012; 116:37-41..

For the Peruvian context, there is an extensive evidence that female generally earn less than their male counterparts ¹⁰10. Ñopo H. Matching as a tool to decompose wage gaps. Rev Econ Stat 2008; 90:290-9.^,1111. del Pozo JM. Has the gender wage gap been reduced during the 'Peruvian Growth Miracle?' A distributional approach. http://departamento.pucp.edu.pe/economia/documento/7859/?lq=en (accessed on 04/Jun/2018).
http://departamento.pucp.edu.pe/economia... ^,1212. Garavito C. Desigualdad en los ingresos: género y lengua materna. In: León Castillo J, Iguiñiz Echeverría J, editores. Desigualdad distributiva en el Perú: dimensiones. Lima: Fondo Editorial de la Pontificia Universidad Católica del Perú; 2011. p. 235-65.. Although some efforts have been developed to describe the health provider wages ¹³13. Taype-Rondan A, Torres-Roman JS, Herrera-Añazco P, Diaz CA, Brañez-Condorena A, Moscoso-Porras MG. Ingresos económicos en médicos peruanos según especialidad: un análisis transversal de la ENSUSALUD 2015. Rev Peru Med Exp Salud Pública 2017; 34:183-91., the empirical evidence of gender wage gap in health occupations is sparse, since data for highly qualified occupations is usually collected in high-income countries ¹⁴14. Magnusson C. The gender wage gap in highly prestigious occupations: a case study of Swedish medical doctors. Work Employ Soc 2016; 30:40-58. and most of specific health surveys are not national representative. In consequence, it is not allowed to propose public policies about this issue. The evidence about gender wage gap in medicine is relevant, as developing countries are exposed to the brain drain phenomenon ¹⁵15. Mullan F. The metrics of the physician brain drain. N Engl J Med 2005; 353:1810-8.^,1616. Okeke EN. Brain drain: do economic conditions "push" doctors out of developing countries? Soc Sci Med 2013; 98:169-78., which means qualified human capital from low income countries must migrate to countries with developed health systems. This loss negatively impacts population wellness ¹⁷17. Taype-Rondán Á, Luna-Porta L, Mendoza Quispe O. El ingreso económico como incentivo para la migración médica: explorando los números. Salud Pública Méx 2016; 58:337-8.^,1818. Aly Z, Taj F. Why Pakistani medical graduates must remain free to emigrate. PLoS Med 2008; 5:e2..

Considering the above, this study is one of the first to explore gender differences in physician labor market in Peru. This study aims to estimate the gender gap in the likelihood of earning high wages for physicians. Furthermore, we decompose this gap to identify the proportion related to differences in individual and labor characteristics (e.g. education, specialty) and the residual proportion related to differences in return to these characteristics, which may be associated to unobservable characteristics and discrimination in the Peruvian labor market as well.

Materials and methods

Study design

Our main data source was the Peruvian National Survey of Users Satisfaction on Health (ENSUSALUD, according to its acronym in Spanish) conducted by the Peruvian National Superintendence of Health and the Peruvian National Institute of Statistics (INEI) ¹⁹19. Superintendencia Nacional de Salud. Encuestas de Satisfacción a nivel Nacional (ENSUSALUD 2015). Lima: Superintendencia Nacional de Salud; 2015. http://portal.susalud.gob.pe/blog/encuestas-de-satisfaccion-a-nivel-nacional-ensusalud-2015/.. This survey is nationally representative and provides a unique dataset that collects information about the wellness of users and professional of health, through six questionnaires. Our analysis specifically relied on the second, focused on academic trajectories and labor activity of health professionals (physicians and nurses).

Population and sample

The ENSUSALUD collects data regarding health providers working on public institutions (Ministry of Health, Public Health Insurance System and Army) and private clinics. This survey follows a two-step stratified probabilistic sampling method, which, in the first step, the analysis units are health service provider institutes; whereas in the second step, the analysis units are a random sample of health professionals for each institute.

For our analysis, we used transversal data of 2014 and 2015 in a pooled dataset. The full sample included around 5,060 health professionals for each year. The sample was restricted to physicians working more than 20 hours per week, in order to control selection bias associated to physicians working fewer hours ⁴4. Jagsi R, Griffith KA, Stewart A, Sambuco D, DeCastro R, Ubel PA. Gender differences in the salaries of physician researchers. JAMA 2012; 307:2410-7.. Finally, we converted a given set of characteristics in discrete variables, corresponding to the applying method ¹⁰10. Ñopo H. Matching as a tool to decompose wage gaps. Rev Econ Stat 2008; 90:290-9..

Variables

For each year, the questions of the survey were all the same. The main variable was the monthly earning, in order to overcome the selection bias related to alternative measure ²⁰20. Baker LC. Differences in earnings between male and female physicians. N Engl J Med 1996; 334:960-4.. The monthly earning had been reported from the following question: “Considering all your paid activities, what is your level of monthly earning?”. This discrete variable reported earning ranking from 1,000 to 5,000 PEN per month. For our analysis, this variable had been modified in a dummy variable, corresponding to “earning < 5,000” and “earning > 5,000”.

Several characteristics are used as controls relies on previous studies ¹³13. Taype-Rondan A, Torres-Roman JS, Herrera-Añazco P, Diaz CA, Brañez-Condorena A, Moscoso-Porras MG. Ingresos económicos en médicos peruanos según especialidad: un análisis transversal de la ENSUSALUD 2015. Rev Peru Med Exp Salud Pública 2017; 34:183-91.^,2121. Rizzo JA, Blumenthal D. Physician labor supply: do income effects matter? J Health Econ 1994; 13:433-53.. Among sociodemographic characteristics, we included dummy variables for ranges of age by tertiles; a dummy variable indicating if the person is married or not; a dummy variable indicating if the person lives with parents or not; and a dummy variable identifying if the person has been sampled in 2014.

Among professional variables, we included a dummy variable identifying the presence of medical specialty or not; and a dummy variable indicating if the person working in the public sector (Ministry of Health, Public Health Insurance System, and Army).

Finally, among labor variables, we included dummy variables for years of labor experience in health sector; dummy variables for ranges of monthly working hours by tertiles, calculated adding all occupations; and dummy variables for the number of occupations, based on questions about working in other institution, working in health institute and working as professor.

Statistical analysis

All analysis was calculated applying survey sampling procedures of ENSUSALUD, that consider the complex sampling design, using Stata (https://www.stata.com). Descriptive analysis was conducted by gender, the t-test was applied for continuous variables and Pearson’s chi-squared test for discrete variables. The gender wage gap analysis was performed applying a novel decomposition method.

Our study used the non-parametric decomposition proposed by Ñopo ¹⁰10. Ñopo H. Matching as a tool to decompose wage gaps. Rev Econ Stat 2008; 90:290-9., which differs from the seminal Blinder-Oaxaca decomposition ²²22. Oaxaca R. Male-female wage differentials in urban labor markets. Int Econ Rev (Philadelphia) 1973; 14:693-709.^,2323. Blinder AS. Wage discrimination: reduced form and structural estimates. J Hum Resour 1973; 8:436-55. in some aspects. First, it does not require the estimation of wage equations to measure the wage gap. Second, it allows restricting the analysis of the wage gap only for individuals with comparable characteristics (into the common support), applying matching techniques. Whether include individual with incomparable characteristics, it might overestimate the unexplained component of the wage gap ¹⁰10. Ñopo H. Matching as a tool to decompose wage gaps. Rev Econ Stat 2008; 90:290-9.. The decomposition method is expressed as follows:

The left side of the equation represents the raw gap in likelihood to earn highly paid wage. The first component on the right side (Δ₀) is the part that cannot be explained by differences in observable characteristics. This component is a combination of unobservable characteristics and discrimination factor associated with the gap. The second and third components (Δ_C and Δ_T , respectively) are the gap portion caused by the presence of male and female physicians with incomparable characteristics. Thus, if all individuals present same characteristics (full sample matching), these components must vanish. Finally, the fourth component (Δ_X ) is the part explained by differences in observable characteristics between male and female physicians (e.g. differences in age groups, presence of medical specialty, labor experience).

Based on previous studies identifying the determinants of physician wage equations ¹³13. Taype-Rondan A, Torres-Roman JS, Herrera-Añazco P, Diaz CA, Brañez-Condorena A, Moscoso-Porras MG. Ingresos económicos en médicos peruanos según especialidad: un análisis transversal de la ENSUSALUD 2015. Rev Peru Med Exp Salud Pública 2017; 34:183-91.^,2121. Rizzo JA, Blumenthal D. Physician labor supply: do income effects matter? J Health Econ 1994; 13:433-53., three specifications were used. In the first one, we only included the aforementioned sociodemographic characteristics: ranges of age, marital status, living with parents and reference years of the sample. In the second, we included the former group of variables and the professional characteristics: medical specialty and work in public sector. Finally, in the third, we added the labor market characteristics to the previous model: labor experience in health sector, ranges of monthly working hours and dummy variables for extra occupations.

Ethical aspects

The data was obtained from the INEI (http://iinei.inei.gob.pe/microdatos/). The interviewers of ENSUSALUD carried out an oral consent with any respondents, indicating that the survey was anonymous and voluntary. The survey only collects information of volunteer respondents.

Results

Descriptive statistics

For each year of our analysis, individuals that did not report full inquiry, income and working less than 20 hours per week were excluded. Our final sample reported 4,282 physicians, including 3,219 males (75.2%) and 1,063 females (24.8%). Table 1 presents descriptive statistics by gender. The preliminary analysis of monthly earnings level shows a higher proportion of male physician earning more than 5,000 PEN comparing with their female counterpart (p < 0.01). According to sociodemographic characteristics, 52% of female physicians and 33% of male physicians are 24 to 40 years old (p < 0.001), 56% of female physicians and 77% of male physicians are married (p < 0.001), 85% of female physicians and 82% of male physicians live with parents (p < 0.01). Regarding professional characteristics, 56% of female physicians and 69% of male physicians have a medical specialty (p < 0.001) Moreover, regarding labor characteristics, female physicians have on average 14 years of experience and male physicians have on average 18 years of experience (p < 0.001), 30% of female physicians and 20% of male physicians work between 90-160 hours per month. In contrast, 28% of female physicians and 40% of male physicians work more than 240 monthly hours per month (p < 0.001). Finally, 54% of female physicians and 37% male physicians reported one occupation (p < 0.001).

Non-parametric wage decomposition

Table 2 presents the measure of the gender gap in the likelihood of earning a high wage and for each of components. Under the different specifications of the model, the proportion of matching people (into the common support) is only significantly reduced in the third specification, being 1,038 male physicians (around 32%) and 605 female physicians (around 57%). As argued before, this suggests heterogeneity in characteristics related to labor market between male and female physicians. Indeed, descriptive statistics show the mismatch among male and female physicians, which is related to higher level in labor characteristics usually for male physicians (e. g. more years of labor experience on the health sector, more working hours per month).

In general, we observe the gap in the likelihood of earning high wages is 81% for male physicians. Using different specifications, the gap decomposition shows that the explained component (Δ_X ) is increasing (from 18 to 25%) by differences in observable characteristics into the common support, whereas the unexplained component (Δ₀) is decreasing (from 62 to 47%). Moreover, the components associated to the non-comparability between groups (Δ_C and Δ_T ) are significant only for the last specification (around 9% in total). In all cases, the higher proportion of the gap is associated to the unexplained component.

Additionally, we analyze the measurement of the gender gap in the likelihood of earning a high wage for some specific characteristics. The Figure 1 shows the unexplained gender gap by ranges of age, marital status, categories of monthly working hours and the presence of medical specialty. The points represent the raw unexplained gap for each characteristic, the upper and lower extremes of the whisker correspond to the maximum and minimum of a 95% confidence interval for the unexplained gender gap. According to monthly working hours, the proportion of the gap presents higher heterogeneity. The unexplained gender gap for physicians that work fewer monthly hours is mostly higher than physicians working more hours. According to the presence of a medical specialty, the variation of this gap is significant. We observe that the unexplained gender gap is higher for physicians without specialty; moreover, analyzing the marital status, it is significant only for married physicians. Finally, analyzing age categories, the unexplained gender gap varies slightly. In all of cases, a higher gap in the likelihood of earning high wages might be associated to higher levels of the gap, since previous results showed that it is its main proportion.

Discussion

Our study contributes to the sparse literature in developing countries regarding differences in payment for prestigious occupation, such as the physician. Our main results indicated that male physicians usually have an 81% higher likelihood of earning high salaries comparing with their female counterparts. Furthermore, the main proportion of this gap is associated to the unexplained component (among 57% and 77%, according to the model specification).

These results are consistent with previous studies performed in developed countries that examine differences on labor outcomes, specifically in payment issues for physicians by gender ⁷7. Bashaw DJ, Heywood JS. The gender earnings gap for US physicians: has equality been achieved? Labour 2001; 15:371-91.^,99. Esteves-Sorenson C, Snyder J. The gender earnings gap for physicians and its increase over time. Econ Lett 2012; 116:37-41.^,1414. Magnusson C. The gender wage gap in highly prestigious occupations: a case study of Swedish medical doctors. Work Employ Soc 2016; 30:40-58.. As it is common, the higher proportion of the gender gap in the likelihood to earn high salaries is accounted for the unexplained component. Therefore, it is necessary to determine potential mechanisms that may explain the residual gap or whether the unexplained component corresponds to prejudice and discrimination against females on the physician labor market in Peru.

Based on the quantitative analysis, we identify some mechanisms that may explain these differences. First, the physician specialty may influence on labor return; moreover, female physicians present higher opportunity cost regarding specialty choice ²⁴24. Essary AC, Green EP, Gans DN. Compensation and production in family medicine by practice ownership. Health Serv Res Manag Epidemiol 2016; 3:2333392815624111.^,2525. Chen MK, Chevalier JA. Are women overinvesting in education? Evidence from the medical profession. J Hum Cap 2012; 6:124-49.. Even, female physicians, holding a specialty, are concentrated in the lower remunerative scale, such as pediatric and practice medicine ²⁶26. LaPierre TA, Hill SA, Jones EV. Women in medicine. In: Connerley ML, Wu J, editors. Handbook on well-being of working women. Dordrecht: Springer; 2016. p. 263-82.. This may be associated to intrinsically patient preferences to receive some services by male physicians ³3. Theurl E, Winner H. The male-female gap in physician earnings: evidence from a public health insurance system. Health Econ 2011; 20:1184-200.^,99. Esteves-Sorenson C, Snyder J. The gender earnings gap for physicians and its increase over time. Econ Lett 2012; 116:37-41.. Second, the preference for more flexible labor schedule is determinant. Female physicians perform a trade-off between less working hours and more hours spent in family and children care ²⁵25. Chen MK, Chevalier JA. Are women overinvesting in education? Evidence from the medical profession. J Hum Cap 2012; 6:124-49.^,2626. LaPierre TA, Hill SA, Jones EV. Women in medicine. In: Connerley ML, Wu J, editors. Handbook on well-being of working women. Dordrecht: Springer; 2016. p. 263-82.. The flexibility in working hours is generally related to penalty in the wage return ²⁷27. Sasser AC. Gender differences in physician pay tradeoffs between career and family. J Hum Resour 2005; 40:477-504.^,2828. Gaiaschi C. Premiums and penalties among physicians in Italy: how gender affects the combined impact of marital and parental status on pay. Polis 2017; 31:97-126.. Third, the presence of unobservable characteristics may be relevant. It is possible to identify a proportion of this unexplained gender gap related to these differences, such as social abilities.

Finally, a commonly factor analyzed in this kind of studies may be related to discrimination against women ⁷7. Bashaw DJ, Heywood JS. The gender earnings gap for US physicians: has equality been achieved? Labour 2001; 15:371-91.. The traditional analysis of the discrimination phenomenon is related to differences in specific outcome among individuals that have same characteristics ²⁹29. Becker GS. The economics of discrimination. Chicago: University of Chicago Press; 2010.. However, recent literature pays attention to the consequences of differences in perception, due to the stereotypes embedded in the society ³⁰30. Bertrand M, Goldin C, Katz LF. Dynamics of the gender gap for young professionals in the financial and corporate sectors. Am Econ J Appl Econ 2010; 2:228-55.. This kind of discrimination is particularly relevant for health sector ²⁷27. Sasser AC. Gender differences in physician pay tradeoffs between career and family. J Hum Resour 2005; 40:477-504.. Public policy proposals to reduce these discrimination issues are generally designed in a long-term perspective, so we have identified some short-term policies, such as informative campaign of the advantages related to health specialty traditionally occupied by male or applying a regulation about standard working hourly paid.

The study showed some limitations. First, the wage has been measured as a discrete variable, excluding exhaustive analysis of the gender wage gap ³¹31. Fortin N, Lemieux T, Firpo S. Decomposition methods in economics. In: Ashenfelder O, Card D, editors. Handbook of labor economics. v. 4, part A. Amsterdam: Elsevier; 2011. p. 1-102.. Moreover, it does not allow the analysis of the gender wage gap distribution and its problems related to differences in higher (glass ceiling) and lower (sticky floor) levels of wages ¹¹11. del Pozo JM. Has the gender wage gap been reduced during the 'Peruvian Growth Miracle?' A distributional approach. http://departamento.pucp.edu.pe/economia/documento/7859/?lq=en (accessed on 04/Jun/2018).
http://departamento.pucp.edu.pe/economia... ^,3232. Carrillo P, Gandelman N, Robano V. Sticky floors and glass ceilings in Latin America. J Econ Inequal 2014; 12:339-61.. Second, the personal health data collected about the sociodemographic characteristics is scarce, with consequences on the model about omitted variable bias. Third, studies usually focus on the relevance of physician specialty ⁹9. Esteves-Sorenson C, Snyder J. The gender earnings gap for physicians and its increase over time. Econ Lett 2012; 116:37-41.. Nevertheless, even though the survey allows distinguishing the presence of specialty, it does not indicate its category (e.g. Pediatrics, general and internal medicine). Fourth, the issue concerns the dimensionality commonly addressed in matching methods, which implies bias using many control variables ³³33. DiNardo J, Fortin NM, Lemieux T. Labor market institutions and the distribution of wages, 1973-1992: a semiparametric approach. Econometrica 1996; 64:1001-44.. Finally, the survey has been implemented since 2014, for this reason we could not realize a longitudinal analysis of the gap.

For developing countries, there is extensively evidence that female usually earn less than their male counterparts ¹⁰10. Ñopo H. Matching as a tool to decompose wage gaps. Rev Econ Stat 2008; 90:290-9.^,3232. Carrillo P, Gandelman N, Robano V. Sticky floors and glass ceilings in Latin America. J Econ Inequal 2014; 12:339-61.^,3434. Canelas C, Salazar S. Gender and ethnic inequalities in LAC countries. IZA Journal of Labor & Development 2014; 3:18.. However, the empirical evidence about gender wage gap in health occupations is scarce ¹³13. Taype-Rondan A, Torres-Roman JS, Herrera-Añazco P, Diaz CA, Brañez-Condorena A, Moscoso-Porras MG. Ingresos económicos en médicos peruanos según especialidad: un análisis transversal de la ENSUSALUD 2015. Rev Peru Med Exp Salud Pública 2017; 34:183-91.^,3535. Lavado P, Martínez J, Yamada G. Calidad de la educación superior y desigualdad en los retornos en el Perú, 2012. Lima: Asociación Peruana de Economía; 2015.. Further, the main contribution of our study is the application of a novel methodology. The non-parametric decomposition proposed by Ñopo ¹⁰10. Ñopo H. Matching as a tool to decompose wage gaps. Rev Econ Stat 2008; 90:290-9. overcome some drawbacks from traditional decomposition method, for example, it does not require the estimation of wage equations to measure the wage gap and restricting the analysis exclusively for male and female physicians with similar characteristics, which generate estimates of lower components related to bias.

In summary, our study has evidenced labor barriers against female physicians, concerning the likelihood to earn high wages and how these differences may be associated to unexplained and discrimination factors. This context is discouraging; however, throughout our study, we explain some mechanism to overcome, reducing labor disparities between male and female physicians.

References

Publication Dates

History