Excerpts from Pager, Western, Bonikowski (2008)

Transcription

1 Excerpts fro Pager, Western, Bonikowski (2008) In one case, for exaple, Zuri, an African Aerican tester, reports his experience applying for a job as a warehouse worker: The original woan who had herded us in told us that when we finished filling out the application we could leave because there s no interview today, guys! When I ade it across the street to the bus stop the woan who had collected our copleted applications pointed in the direction of Sion, Josue and yself [the three test partners] otioning for us to return. All three of us went over. She looked at e and told e she needed to speak to these two and that I could go back. Zuri returned to the bus stop, while his white and Latino test partners were both asked to coe back at 5p that day to start work. Sion, the white tester, reports, She said she told the other people that we needed to sign soething that that s why she called us over so as not to let the know she was hiring us. She seeed pretty concerned with not letting anyone else know.

2 In one case, for exaple, the three testers inquired about a sales position at a retail clothing store. Joe, one of our African Aerican testers, reports: [The eployer] said the position was just filled and that she would be calling people in for an interview if the person doesn t work out. Josue, his Latino test partner, was told soething very siilar: She infored e that the position was already filled, but did not know if the hired eployee would work out. She told e to leave y resue with her. By contrast, when Sion, their white test partner, applied last, his experience is notably different: I asked what the hiring process was if they re taking applications now,interviewing, etc. She looked at y application. You can start iediately? Yes. Can you start toorrow? Yes. 10 a.. She was very friendly and introduced e to another woan (white, 28) at the cash register who will be training e.

3 Table 1. Percentage of positive responses and race differences, by level of personal contact White Latino Black Race Differences Subsaple (W) (L) (B) W / L W / B L / B Total No personal contact Personal contact White felon Latino Black Race Differences Subsaple (Wf) (L) (B) Wf / L Wf / B L / B Total No personal contact Personal contact Note: Personal contact varies across testers within teas. Tests involving personal contact represent 56% by white testers, 49% by Latino testers, and 61% by black testers in the first tea (N=171); 44% of white testers, 45% of Latino testers, and 31% of black testers in the second tea (N=170). Source: Pager, Western, Bonikowski (2006)

4 Table 1. Job Channeling by Race original job title suggested job Blacks channeled down Server Busser (324) Counter person Dishwasher/porter (102) Server Busboy (189) Assistant anager Entry fast food position (258) Server Busboy/runner (269) Retail sales Maintenance (399) Counter person Delivery (176) Sales Stockboy (831) Sales Not specified (a) Hispanics channeled down Server Runner (199) Sales Stock (2) Stea cleaning Exterinator (79) Counter person Delivery (176) Sales Stock person (503) Whites channeled down Server Busboy (192) Hispanics channeled up Carwash attendant Manager (1058) Warehouse worker Coputer/office (1001) Whites channeled up Line Cook Waistaff (254) Mover Office / Telesales (784) Dishwasher Waistaff (858) Driver Auto detailing (948) Kitchen job Front of the house job (5) Receptionist Copany supervisor (347) (a) eployer told tester sales ight not be right for you Note: nubers in parentheses refer to eployer ID codes.

5 VOL. 94 NO. 4 BERTRAND AND MULUINATHAN: RACE IN THE LABOR MARKET 997 Saple: All sent resues Percent callback Percent callback for Percent difference for White naes African-Aerican naes Ratio (D-value) Chicago Boston Feales Feales in adinistrative jobs Feales in sales jobs Males Notes: The table reports, for the entire saple and different subsaples of sent resues, the callback rates for applicants with a White-sounding nae (colun 1) an an African-Aerican-sounding nae (colun 2), as well as the ratio (colun 3) and difference (colun 4) of these callback rates. In brackets in each cell is the nuber of resues sent in that cell. Colun 4 also reports the p-value for a test of proportion testing the null hypothesis that the callback rates are equal across racial groups.

6 1012 THE AMERICAN ECONOMIC REVIEW SEPTEMBER 2004 TABLE AI-FIRST NAMESUSEDIN EXPERIMENT White feale Nae L(WI/L(BI Percevtion White Allison Anne Carrie Eily Jill Laurie Kristen Meredith Sarah Y2 Fraction of all births: 3.8 percent White ale Nae L(W)k(B) Perception White Brad Brendan Geoffrey Greg Brett Jay Matthew Neil Todd cc a a Fraction of all births: 1.7 percent African-Aerican feale Nae L(BI/L(WI Perce~tion Black Aisha Ebony Keisha Kenya Lakisha Latonya Latoya Taika 284 Tanisha 32 Fraction of all births: 7.1 percent African-Aerican ale Nae L(B)/L(W) Perception Black Darnell Haki Jaal 257 Jeraine 90.5 Karee Leroy 44.5 Rasheed Treayne Tyrone 62.5 Fraction of all births: 3.1 percent Notes: This table tabulates the different first naes used in the experient and their identifiability. The first colun reports the likelihood that a baby born with that nae (in Massachusetts between 1974 and 1979) is White (or African-Aerican) relative to the likelihood that it is African-Aerican (White). The second colun reports the probability that the nae was picked as White (or African-Aerican) in an independent field survey of people. The last row for each group of naes shows the proporrion of all births in that race group that these naes account for.

7 VOL. 94 NO. 4 BERTRAND AND MULLAINATHAN: RACE IN THE LABOR MARKET TABLE&-EFFECT OF APPLICANT'S ADDRESS ON LIKELIHOOD OF CALLBACK 1003 Dependent Variable: Callback Duy Fraction college or Zip code characteristic: Fraction Whites ore Log(per capital incoe) Zip code characteristic Zip code characteristic* African-Aerican nae African-Aerican nae (0.012) (0.016) (0.022) (0.031) (0.007) (0.010) (0.024) (0.048) (0.015) (0.015) (0.013) (0.152) Notes: Each colun gives the results of a probit regression where the dependent variable is the callback duy. Reported in the table is the estiated arginal change in probability. Also included in coluns 1, 3, and 5 is a city duy; also included in coluns 2,4, and 6 is a city duy and a city duy interacted with a race duy. Standard errors are corrected for clustering of the observations at the eployent-ad level.

8 1004 THE AMERICAN ECONOMIC REVIEW SEPTEMBER 2004 TABLE7-EFFECT OF JOB REQUIREMENT AND EMPLOYER CHARACTERISTICS ON RACIAL DIFFERENCES IN CALLBACKS Saple ean Marginal effect on callbacks Job requireent: (standard deviation) for African-Aerican naes Any requireent? (Y = 1) (0.41) (0.015) Experience? (Y = 1) (0.49) (0.013) Coputer skills? (Y = 1) (0.50) (0.013) Counication skills? (Y = 1) (0.33) (0.015) Organization skills? (Y = I) (0.26) (0.029) Education? (Y = 1) (0.31) (0.017) Total nuber of requireents (0.93) (0.006) Saple ean Marginal effect on callbacks Eployer characteristic: (standard deviation) for African-Aerican naes Equal opportunity eployer? (Y = 1) (0.45) (0.012) Federal contractor? (Y = 1) (N = 3,102) (0.32) (0.016) Log(ep1oyent) (N = 1,690) (1.74) (0.005) Ownership status: (N = 2,878) Privately held (0.019) Publicly traded (0.015) Not-for-profit (0.042) Fraction African-Aericans in eployer's zip code (N = 1,918) (0.15) (0.062) Notes: Saple is all sent resues (N = 4,870) unless otherwise specified in colun 1. Colun 2 reports eans and standard deviations (in parentheses) for the job requireent or eployer characteristic. For ads listing an experience requireent, 50.1 percent listed "soe," 24.0 percent listed "two years or less," and 25.9 percent listed "three years or ore." For ads listing an education requireent. 8.8 percent listed a high school degree, 48.5 percent listed soe college, and 42.7 percent listed at least a four-year college degree. Colun 3 reports the arginal effect of the job requireent or eployer characteristic listed in that row on differential treatent. Specifically, each cell in colun 3 corresponds to a different probit regression of the callback duy on an African-Aerican nae duy, a duy for the requireent or characteristic listed in that row and the interaction of the requireent or characteristic duy with the African-Aerican nae duy. Reported in each cell is the estiated change in probability for the interaction ter. Standard errors are corrected for clustering of the observations at the eployent-ad level.

9 1008 THE AMERICAN ECONOMIC REVIEW SEPTEMBER 2004 TABLE 8---CALLBACK RATE AND MOTHER'S EDUCATION BY FIRST NAME White feale African-Aerican feale Nae Percent callback Mother education Nae Percent callback Mother education Eily Anne Jill Allison Laurie Sarah Meredith Carrie Kristen Aisha Keisha Taika Lakisha Tanisha Latoya Kenya Latonya Ebony Average 91.7 Average Overall 83.9 Overall Correlation (p = 0.404) Correlation (p = 0.309) White ale African-Aerican ale Nae Percent callback Mother education Nae Percent callback Mother education Todd Rasheed Neil Treayne Geoffrey Karee Brett Darnell Brendan Tyrone Greg Haki Matthew Jaal Jay Leroy Brad Jeraine Average 91.7 Average Overall 83.5 Overall Correlation (p = 0.949) Correlation (p = 0.120) Notes: This table reports, for each first nae used in the experient, callback rate and average other education. Mother education for a given first nae is defined as the percent of babies born with that nae in Massachusetts between 1970 and 1986 whose other had at least copleted a high school degree (see text for details). Within each sextrace group, first naes are ranked by increasing callback rate. "Average" reports, within each race-gender group, the average other education for all the babies born with one of the naes used in the experient. "Overall" reports, within each race-gender group, average other education for all babies born in Massachusetts between 1970 and 1986 in that race-gender group. "Correlation" reports the Spearan rank order correlation between callback rate and other education within each race-gender group as well as the p-value for the test of independence.

10 Source: Bertrand and Duflo (2016) Table 1: Labor arket correspondence studies Paper Country CVs / apps Vacancies Effect (Call-back ratio) Theory Galarza and Yaada (2014) Peru 4,820 1,205 White-to-indigenous ratio: 1.8 No Trait: Ethnicity; Attractiveness Low attractiveness hurts white feales Eriksson and Rooth (2014) Sweden 8,466 - Eployed to long-ter uneployed: 1.25 No Trait: Uneployent duration Bloaert, Coenders, and van Tubergen (2014) Netherlands Dutch-to-foreign: 1.62 (unconditional No Trait: Arabic nae ratio). No difference, if views held fixed Nunley, Pugh, Roero, and Seals (2014) US 9,396 - White-to-black: 1.18 (unconditional) Inconsistent with statistical Trait: Race discriination, consistent with taste-based discriination Ghayad (2013) US Eployed-to-uneployed: 1.47 No Trait: Uneployent duration Bartoš, Bauer, Chytilová, and Matějka (2013) Czech Rep. 274 (Czech R.) - Czech-to-Vietnaese: 1.34 Consistent with attention Trait: Ethnicity (Roa, Asian, Turkish) and Gerany 745 (Ger.) Lower requests for CVs if candidate discriination is Turkish 14 Wright, Wallace, Bailey, and Hyde (2013) US 6,400 1,600 White-to-Musli: 1.58 Consistent with theoretical Trait: Religion / ethnicity odels of secularization and cultural distate theory Kroft, Lange, and Notowidigdo (2013) US (largest ,040 1 log point change in uneployent No Trait: Uneployent duration 100 MSAs) duration: 4.7 percentage points lower call-back probability Baert, Cockx, Gheyle, and Vandae (2013) Belgiu Fleish-to-Turkish: 1.03 to 2.05, No Trait: Nationality (Turkish-sounding nae) depending on the occupation Bailey, Wallace, and Wright (2013) US 4,608 1,536 No effect No Trait: Sexual orientation Ahed, Andersson, and Haarstedt (2013) Sweden 3,990 - Heterosexual-to-hoosexual (ale): 1.14 No Trait: Sexual orientation Heterosexual-to-hoosexual (feale): 1.22 Acquisti and Fong (2013) US Christian-to-Musli: 1.16 No Traits:Sexual orientation and religion Patacchini, Ragusa, and Zenou (2012) Italy 2,320 - Heterosexual-to-Hoosexual: 1.38 No Traits: Sexual orientation and attractiveness Kaas and Manger (2012) Gerany 1, Geran-to-Turkish: 1.29 Consistent with statistical Trait: Iigrant (race/ethnicity) (if no reference letter is included) discriination Maurer-Fazio (2012) China 21,592 10,796 Han-to-Mongolian: 1.36 No Trait:Ethnicity Han-to-Tibetan: 2.21

11 Source: Bertrand and Duflo (2016)...Continued fro previous page Paper Country CVs / apps Vacancies Effect (Call-back ratio) Theory Jacqueet and Yannelis (2012) US English-to-foreign naes: 1.41 Consistent with patterns Trait: Race / Nationality English-to-Black naes: 1.46 of ethnic hoophily Ahed, Andersson, and Haarstedt (2012) Sweden year old-to-46 year old: 3.23 No Trait: Age Oreopoulos (2011) Canada English nae-to-iigrant: ranged No Trait: Nationality (and race) fro 1.39 to 2.71 (against Indian Pakistani and Chinese applicants) Carlsson (2011) Sweden 3,228 1,614 Feale-to-Male: 1.07 No Trait: Gender Booth, Leigh, and Varganova (2011) Australia Above White-to-Italian: 1.12 No Trait: Ethnicity White-to-Chinese: 1.68 Booth and Leigh (2010) Australia 3,365 - Feale-to-ale: 1.28 No Trait: Gender (feale-doinated professions) 15 Riach and Rich (2010) UK 1, favoring younger candidates No Trait: Age Rooth (2009) Sweden 1, Non-obese/attractive-to-obese/unattractive: No Trait: Attractiveness/Obesity ranged fro 1.21 to 1.25 (but higher for soe occupations) McGinnity, Nelson, Lunn, and Quinn (2009) Ireland , 2.07, 2.44 in favor of Irish and against No Trait: Nationality / race Asians, Gerans and Africans respectively Banerjee, Bertrand, Datta, and Mullainathan (2009) India 3, Upper Caste-to-Other Backward Castes: 1.08 Traits: Caste and religion (software jobs, insignificant), 1.6 (call-center jobs) No Lahey (2008) US App. 4,000 - Young-to-older: 1.42 No Trait: Age Petit (2007) France Ranged fro 1.13 to 2.43 against 25-year-old, No Traits: Age, gender, nuber of children childless woen Bursell (2007) Sweden 3,552 1,776 Swedish-to-foreign naes: 1.82 Inconsistent with statistical Trait: Ethnicity discriination Bertrand and Mullainathan (2004) US 4, White-to-African-Aerican: 1.5 No Trait: Race (1.22 for feales in sales jobs) Jolson (1974) US White-to-black: 4.2 for selling positions No Trait: Race and religion

12 Source: Bertrand and Duflo (2016) Table 2: Rental arket papers Study Country Inquiries Effect Theory Carlsson and Eriksson (2014) Sweden 5,827 Swedish-to-Arabic (feales): 1.37 No Trait: Minority status (Arabic nae) Swedish-to-Arabic (ales): 1.62 Ewens, Tolin, and Wang (2014) US 14,237 White-to-Black: 1.19 Consistent with statistical Trait: Race discriination, inconsistent with taste-based discriination Bartoš, Bauer, Chytilová, and Matějka (2013) Czech Republic 1,800 Czech-to-inority: 1.27 (site available), Consistent with attention Trait: Minority status (Roa or Asian nae) and Gerany 1.9 (pooled Asian and Roa naes) discriination Hanson and Hawley (2011) US 9,456 White-to-African Aerican: 1.12 Consistent with statistical Trait: Race (varied by neighborhood and unit type) discriination 21 Baldini and Federici (2011) Italy 3,676 Italian-to-East European: 1.24 No Trait: Iigrant status; Language ability Italian-to-Arab: 1.48 Ahed, Andersson, and Haarstedt (2010) Sweden 1,032 Swedish-to-Arab/Musli: 1.44 (no No Trait: Minority status (Arabic nae) inforation), 1.24 (detailed inforation about the applicant) Bosch, Carnero, and Farré (2010) Spain 1,809 Spanish-to-Moroccan: 1.44 (no No Trait: Iigrant status inforation), 1.19 (with positive inforation) Ahed and Haarstedt (2009) Sweden 408 Straight-to-gay: 1.27 No Trait: Sexual orientation Ahed and Haarstedt (2008) Sweden 1,500 Swedish-to-Arab ale: 2.17 No Trait: Iigrant (race/ethnicity/religion) Carpusor and Loges (2006) US (Los Angeles County) 1,115 White-to-Arab: 1.35 No Trait: Race / Ethnicity (Arab, African-Aerican) White-to-Black: 1.59, conditional on hearing back, 1.98 unconditional

13 Doleac and Stein (2013) 2013] THE VISIBLE HAND F473 (a) (b) (c) (d) (e) (f) (g) (h) (i) Fig. 1. Advertiseent Photographs Note. These photographs have been slightly scaled down fro the size included in our advertiseents.

14 Doleac and Stein (2013) 2013] THE VISIBLE HAND F479 Table 2 Key Outcoe Averages by Advertiseent Type White Black Tattoo Total Preaturely reoved Nuber of responses Nuber of non-scas Nuber of offers Received 1 offer Indicators of trust in responses (given 1 non-sca response) Includes nae Uses polite language Includes personal story Offer aount Mean offer Best offer Offer aount (given 1 offer) Mean offer Best offer Reaction to delivery proposal (given delivery proposed) Sca/payent concern No response Other Prefer to wait Willing to ship ipod shipped Notes. Mean values are reported. Observations are weighted by state population/nuber of advertiseents posted in each state.

15 Source: O'Neil and O'Neil (2005) Table 10 Means and Partial Regression Coefficients of Explanatory Variables 1) fro Separate NLSY Log Wage Regressions for Men and Woen Ages in 2000 Means Feale Male Feale Male M2 M4 M2 M4 Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Race Hispanic (0,1) Black (0,1) Education and skill level <10 yrs yrs (no diploa or GED) * HS grad (diploa) HS grad (GED) Soe college BA or equiv. degree MA or equiv. degree Ph.D or prof. Degree AFQT percentile score (x.10) L.F. withdrawal due to faily responsibilities (0,1) Lifetie Work Experience Weeks worked in civilian job since age Weeks worked in ilitary since Weeks PT total weeks workd since age Eployent type Gov't eployer (0,1) Non-profit eployer (0,1) OCC. Characteristics of Person's 3-digit OCC. SVP required in occup. (onths) (DOT) Hazards (0,1) (DOT) Fues (0,1) (DOT) Noise (0,1) (DOT) Strength (0,1) (DOT) Weather extree (0,1) (DOT) Prop. using coputers (CPS) Prop. using coputer for analysis (CPS) Prop. using coputer for word proc. (CPS) Relative rate of transition to uneployent Relative rate of transition to OLF % feale in OCC. X 0.1. (CPS ORG) Adj. R-Square Dependent ean (Log Hourly Wage) Saple size ) Model also controls for age, central city, MSA, region, and occupation issing. * Reference group. Source: National Longitudinal Survey of Youth (NLSY79) erged with easures of occupational characteristics (3-digit level) fro the Septeber 2001 CPS, the March CPS, the CPS ORG, and the Dictionary of Occupational Titles (1991).

16 Source: O'Neil and O'Neil (2005) Table 11 Gender Wage Gap: Decoposition Results (NLSY, 2000) Using ale coefficients Using feale coefficients Log Wage Gap (Male-Feale) Attributable to: M1 M2 M3 M4 M1 M2 M3 M4 Age, race, region, central city, MSA AFQT Education level L.F. withdrawal due to faily responsibilities Lifetie work experience Nonprofit, governent Occupational characteristics: Investent related SVP (Specific Vocational Preparation) Coputer usage Copensating differences Disaenities (physical) Uneployent risk; labor force turnover TYP: % feale in occupation Unadjusted log wage gap Total explained by odel Unexplained log wage gap Unadjusted hourly wage ratio (Feale/Male) : Adjusted hourly wage ratio (Feale/Male) : Note: Decoposition results shown are derived fro results of separate regressions for en and woen. See Table 10 for variable eans and coefficients using Model 2 and 4. Wage ratios are based on the exponentiated log hourly wage. Source: National Longitudinal Survey of Youth (NLSY79) erged with easures of occupational characteristics (3-digit level) fro the Septeber 2001 CPS, the March CPS, the CPS ORG, and the Dictionary of Occupational Titles (1991).

17 Source: O'Neil and O'Neil (2005) Table 4 Means and Partial Regression Coefficients of Explanatory Variables 1) fro Separate Log Wage Regressions for Black, White, and Hispanic MEN Ages in 2000 (NLSY) Mean White Black Hispanic White Black Hisp. M1 M2 M1 M2 M1 M2 Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Education and skill level <10 yrs yrs (no diploa or GED) * HS grad (diploa) HS grad (GED) Soe college BA or equiv. degree MA or equiv. degree Ph.D or prof. Degree AFQT percentile score (x.10) Lifetie work experience (Year equivalents) Weeks worked in civilian job since age Weeks worked in ilitary since Adj. R-Square Dependent ean (Log Hourly Wage) Saple size ) Model also controls for age, central city, MSA and region. The analysis is restricted to wage and salary workers eployed within the past onth. * Reference group. Source: National Longitudinal Survey of Youth (NLSY79).

18 Source: O'Neil and O'Neil (2005) Table 5 White-Black and White-Hispanic Wage Gaps: Decopositon Results for MEN (NLSY) White-Black Differential White-Hispanic Differential Using black ale coef. Using white ale coef. Using hispanic ale coef. Using white ale coef. M1 M2 M1 M2 M1 M2 M1 M2 Log Wage Gap Attributable to: Age, region, central city, MSA AFQT Education Lifetie work experience Unadjusted log wage gap Total explained by odel Unexplained log wage gap Unadjusted inority/white hourly wage ratio: Adjusted inority/white hourly wage ratio: Note: Decoposition results shown are derived fro results of separate regressions for en ages by race and by odel using NLSY79 data fro the 2000 survey. See Table 4 for variable eans and coefficients. Hourly wages are the exponentiated hourly log wages. Source: National Longitudinal Survey of Youth (NLSY79).

19 Reference Group: Male Coef. Table 4. Gender Wage Gap: Quantile Decoposition Results (NLSY, 2000) 10th percentile 50th percentile 90th percentile A: Raw log wage gap : Q τ [ln(w )]-Qτ[ln(w f )] ( 0.023) ( 0.019) ( 0.026) B: Decoposition Method: Machado-Mata-Melly Estiated log wage gap: Qτ[ln(w )]-Qτ[ln(w f )] ( 0.015) ( 0.016) ( 0.026) Total explained by characteristics ( 0.028) ( 0.027) ( 0.019) Total wage structure ( 0.027) ( 0.024) ( 0.025) C: Decoposition Method: RIF regressions without reweighing Mean RIF gap: E[RIF τ (ln(w ))]-E[RIF τ (ln(w f ))] ( 0.023) ( 0.019) ( 0.026) Coposition effects attributable to Age, race, region, etc ( 0.005) ( 0.004) ( 0.004) Education ( 0.005) ( 0.006) ( 0.01) AFQT ( 0.02) ( 0.004) ( 0.005) L.T. withdrawal due to faily ( 0.021) ( 0.014) ( 0.017) Life-tie work experience ( 0.026) ( 0.014) ( 0.023) Industrial Sectors ( 0.012) ( 0.008) ( 0.011) Total explained by characteristics ( 0.035) ( 0.025) ( 0.028) Wage structure effects attributable to Age, race, region, etc ( 0.426) ( 0.357) ( 0.524) Education ( 0.028) ( 0.031) ( 0.045) AFQT ( 0.03) ( 0.042) ( 0.062) L.T. withdrawal due to faily ( 0.032) ( 0.025) ( 0.032) Life-tie work experience ( 0.148) ( 0.082) ( 0.119) Industrial Sectors ( 0.06) ( 0.046) ( 0.052) Constant ( 0.349) ( 0.323) ( 0.493) Total wage structure ( 0.044) ( 0.028) ( 0.036) Note: The data is an extract fro the NLSY79 used in O'Neill and O'Neill (2006). Industrial sectors have been added to their analysis to illustrate issues linked to categorical variables. The other explanatory variables are age, duies for black, hispanic, region, sa, central city. Bootstrapped standard errors are in parentheses. Means are reported in Table 2.

20 Chart 2.7. The gender wage gap adjusted for the effect of the wage structure a Percentage difference between ale and feale average gross hourly wages, persons aged 20 to 64 years b Hourly wage gap Hourly wage gap adjusted for the reuneration rates of observed characteristics Hourly wage gap adjusted for the whole wage structure % % Netherlands Austria Gerany France Portugal United Kingdo Finland Ireland Denark Belgiu Greece Spain Italy 0 a) The gender wage gap adjusted for cross-country differences in the reuneration rates of observed characteristics is obtained as follows: adjobs log Wi = log Wi X i ( β i β k ),where i indexes countries, k denotes the benchark country, and refer to country averages and differences between en and woen, respectively, W stands for gross hourly wages, X for the vectors of observed characteristics, and β for the vector of estiated coefficients fro the ale wage regressions (cf. Annex Table 2.B.1). The gender wage gap adjusted for crosscountry differences in the whole wage structure is obtained as follows: log W i adj = log Wi X i ( β i β k ) ( ε i η ik ) = X i β k + η ik, where ε stands for the residuals fro the ale wage regressions (defined as the difference between actual and predicted values) and η for the theoretical residuals that would be obtained in country i if it had the sae residual wage structure as country k. b) Countries are ranked by decreasing hourly wage gap adjusted for the whole wage structure. Sources and definitions: See Annexes 2.A and 2.B respectively. OECD EMPLOYMENT OUTLOOK ISBN