WWW.THESES.XLIBX.INFO
FREE ELECTRONIC LIBRARY - Theses, dissertations, documentation
 
<< HOME
CONTACTS



Pages:     | 1 |   ...   | 5 | 6 ||

«IZA DP No. 3067 How General Is Human Capital? A Task-Based Approach Christina Gathmann Uta Schönberg September 2007 Forschungsinstitut zur Zukunft ...»

-- [ Page 7 ] --

We also require a person to be below a certain age when we …rst observe them. This ensures that we can follow them from day one of their entry into the labor market. The age restriction is 19 if the individual has no high school degree (Abitur ), 22 if the individual has a high school degree, but no higher degree, 28 if the individual graduated from a community college (Fachhochschule), and 30 if he graduated from university. Finally, we drop all observations we observe less than a year, with missing education or nationality, and observations with no valid wage or a daily (real) wage below 20 DM during an employment spell. To estimate the returns to di¤erent types of human capital in Section 5, we use a sample of displaced workers. A worker is displaced from his …rm due to plant closure if he left the …rm in the year or one year before the …rm closed down. As a robustness check, we have repeated the analysis restricting the sample to workers who have left the …rm in the year or one or two year before the …rm closed down. The …rst de…nition has the advantage that it includes less workers who have left the …rm voluntarily, for reasons other than plant closure. It has the disadvantage that it may exclude workers that leave the …rm prior to plant closure, anticipating that the …rm may shut down in the future. Both de…nitions give very similar results.

B Blundell and Powell (2004) Control Variable Estimator For the high-skilled, the control variable estimates in Table 9, columns (1) to (4) are based on the semiparametric estimator proposed by Blundell and Powell (2004). This estimator accounts for censoring in addition to endogenous regressors. It does so however at the price of imposing common returns on the observable human capital variables. For simplicity, we drop the subscript o: The model is

–  –  –

where Xit are the endogenous regressors, uit is the scalar error term and ct is the time-dependent censoring point. For notational convenience, we suppress all exogenous regressors here. The reduced

form links the instruments Zit to the endogenous regressors:

–  –  –

where vit is a scalar error term and the unknown coe¢ cient vector with suitable dimensions. Instead of imposing independence between (uit ; vit ) and Zit, the estimator imposes a weaker conditional quantile exclusion restriction

–  –  –

Notes : The table reports the percentage of individuals in the career survey that report performing the type of task in their job. We grouped the 19 different tasks into three task groups (analytical, manual and interactive skills) following Autor et al. (2003) and Spitz (2006). The fraction for main tasks sum to more than 100 percent as around 10 percent reported performing more than one main task. The last two columns show the distribution of task usage for two common occupations: teachers (which exclude university or technical college professors) and baker.

Source : Qualification and Career Survey: 1979, 1985, 1991/2, 1997/8 Table 2: Measuring Distances between Occupations

–  –  –

Notes : The table shows at the top summary statistics of the distance measure as well as the three most similar and distant occupations and their corresponding distance. The distance measure is the angular separation using the 19 different tasks (see Table B1 for a list of tasks) and normalized to vary between 0 and 1. The bottom part of the table shows the three most commonly observed moves in the data by education group and the corresponding distance measure.

Table 3: Summary Statistics of West German Employee Panel

–  –  –

Notes : The table reports summary statistics for the administrative panel data on individual labor market histories and wages from 1975 to

2001. Low skilled workers are those without a vocationaldegree, medium skilled have either a high school or vocational degree and the high skilled have an advanced degree from a technical college or university. Experience, occupation, task and firm tenure are measured from actual spells and exclude periods of unemploymentor out of the labor force. The wage is measured in German Marks at 1995 prices and is subject to right censoring.

Source : Employee Sample (IAB), 1975-2001 Table 4: Observed Moves are More Similar than under Random Mobility

–  –  –

Notes : The table repports selected moments of the distribution of observed occupational moves ("Observed Mobility") and compares it against what we would expect to observe under random mobility ("Random Mobility"). We calculate random mobility as follows: for each mover, we assume that the probability of going to any other occupation in the data is solely determined by the relative size of the target occupation. We then multiply this "random move" with its distance to get the distribution of the distance measure under random mobility. The distance measure is the angular separation, based on 19 tasks. Since all moments of the observed distribution are below those under random mobility, individuals are much more likely to move to similar occupation.

Table 5: Distance of Move Declines with Time in the Labor Market

–  –  –

Notes : The table reports results from wage regressions where the dependent variable is the log daily wages at the target occupation after an occupational move. Results are reported separately by education group. For the lowand medium-skilled, we estimate OLS models. Standard errors in parentheses allow for clustering at the individual level. For the high-skilled, we estimate tobit models, and exclude censored observations at the previous occupation. Standard errors in parentheses are bootstrapped with 100 replications to allow for clustering at the individual level. All specifications include the log daily wage in the last period, actual experience, actual experience squared, year and occupation dummies. Column (1) uses the sample of occupational stayers as a benchmark for comparison. Column (2) repeats the analysis for occupational movers, while Column (3) adds the distance measure as well as the distance measure interacted with the wage last period. The distance measure used is the angular separation based on all 19 tasks. Coefficients with * are statistically significant at the 5 percent level, those with ** at the 1 percent level.





Table 7: Past Occupational Tenure Matters for Wages

–  –  –

Notes : The table reports wage regressions where the dependent variable is the log wages in the target occupation after an occupational move. Estimates are reported for each education group separately. For the low- and medium-skilled, we report results from OLS regressions. Standard errors in parentheses allow for clustering at the individual level. For the high-skilled, we estimate tobit models. Here, standard errors in parentheses are bootstrapped with 100 replications to account for clustering at the individual level. Column (1) in each specification controls for past tenure in the source occupation, experience, experience squared, as well as year and occupation dummies. Column (2) additionally includes the distance measure interacted with past occupational tenure. The distance measure used is the angular separation based on all 19 tasks. Coefficients with * are statistically significant at the 5 percent level, those with ** at the 1 percent level.

Table 8: Returns to Labor Market Skills: Least Squares

–  –  –

Notes : The table reports results from a regression of the log daily wage on general human capital (experience, experience squared), firm tenure, occupation and task tenure. All specifications include year, region and occupation dummies. Panel C estimates tobit models to account for censoring. Specifications in columns (2), (4) and (6) add our measure of task tenure to the specification in columns (1), (3) and (5). Columns (1)-(2) are estimated for the whole sample, columns (3)-(4) on those who have switched firms and columns (5)-(6) on our sample of displaced workers. Standard errors allow for clustering at the individual level. For Panel C, standard errors are bootstrapped with 100 replications to account for clustering at the individual level. Coefficients with ***, **, * are significant at the 1, 5 and 10 percent level respectively.

Table 9: Returns to Labor Market Skills: Control Function

–  –  –

Note : For the medium- and low-skilled (Panel A and B), columns (1) to (4) report the control function estimates for those who have switched firms (columns (1) and (2)) and our sample of displaced workers (columns (3) and (4)). Columns (5) and (6) report first difference estimates using the sample of firm switchers. For the high-skilled (Panel C), columns (1) to (4) report a semiparametric estimator proposed by Blundell and Powell (2004) to account for censoring in addition to endogenous regressors. Columns (5) and (6) show fixed effects estimates using Honore's semiparametric trimming procedure for tobit models. In all specifications, standard errors are bootstrapped with 100 replications and allow for clustering at the individual level. All specifications include year, region and occupation dummies. Note that due to differences in the econometric model, results of the high-skilled are not directly comparable to those of the other two education groups. Coefficients with ***, **, * are significant at the 1, 5 and 10 percent level.

Figure 1: Quarterly Occupation Quit Rate by Time in the Labor Market

–  –  –

Notes: The figure shows the quarterly occupation quit rate by education and time in the labor market (potential experience). Quit rates are defined over the sample of workers who are employed at the beginning of the quarter.

Figure 2: Observed Mobility is More Similar Than Random Mobility

–  –  –

Notes : The figure plots the density of the distance measure under observed and random mobility. We calculate random mobility as follows: for each mover, we assume that the probability of going to any other occupation in the data is solely determined by the relative size of the target occupation. We then multiply this "random move"

with its distance to get the distribution of the distance measure under random mobility. Distance measure:

angular separation, 19 tasks.

Figure 3: Distance of Occupational Moves Declines over Career

–  –  –

Notes: The figure plots the average distance of the occupational move by actual experience. Regressions control for 15 experience dummies, occupation dummies, and time dummies. The decline in the average distance by experience is significant at a 1 % level for all education groups. Distance measure: angular separation based on 19 tasks.

Figure 4a: Correlation of Wages by Distance of Move

–  –  –

Notes: The upper panel plots the impact of the past wage on the current wage by the distance of the occupational move. Regressions control for occupation and time dummies, past wages, ten distance dummies as well as the past wage interacted with the 10 distance dummies. The lower panel plots the impact of past occupation tenure on current wages by the distance of the occupational move. Regressions control for occupation and time dummies, past occupation tenure and past occupation tenure interacted with ten distance dummies.

Table A1: List of Occupations and Task Usage

–  –  –

Notes : The table shows the title of the 64 occupations, the percentage of individuals employed in it and the fraction of individuals that report performing analytical, manual and interactive tasks on their job following the classification of Autor et al (2003). For a description of the tasks underlying the three aggregate task groups, see Table B2.

Source : IAB Employee Sample, matched with Qualification and Career Survey: 1979, 1985, 1991/2, 1997/8.

Table A2: Estimates of Reduced Forms for Control Function Estimator (Table 9)

–  –  –

Notes : The table reports the regression results of the reduced forms for experience, occupational tenure and task tenure used to construct the control function in Table 9, for the low- and medium-skilled. All specifications are estimated on the sample of firm switchers in columns (1) to (3) and on the sample of displaced workers in columns (4) to (6). For each education group, the dependent variable is experience (columns (1) and (4)), occupational tenure (columns (2) and (5)) and task tenure (columns (3) and (6)) respectively. All specifications include occupation, year and region dummies. Coefficients with ***, **, * are significant at the 1, 5 and 10 percent level respectively. See also notes to Table 9.

Table A3: First- and Second-Stage Regression of Control Variable Estimator for Models with Censoring

–  –  –

Notes : For the high-skilled, the table reports the regression of experience, occupational tenure and task tenure on the instruments in columns (7)-(9). In column (10), it reports the results of a quantile regression of the log wage on the instruments and endogenous controls. All specifications include year, occupation and region dummies. See also notes to Table 9 and the Appendix.

Table A4: Impact of Residuals on Wage Equation (Table 9)

–  –  –

Notes : The table reports the coefficients on the residuals and their interaction with the main regressors to control for selection in Table 9. The column number in this table correspond to the column numbers (1) to (4) in Table 9. Standard errors are bootstrapped with 100 replications. The last column reports the p-value of the test for joint significance of the residuals and the interaction terms. See also notes to Table 9.



Pages:     | 1 |   ...   | 5 | 6 ||


Similar works:

«Policies and Procedures Manual This is a CLIA-waived system. Approval date Approved by_Reviewed/Revised: Date By whom Title Roche Diagnostics 9115 Hague Road Indianapolis, IN 46256 www.poc.roche.com COAGUCHEK is a trademark of Roche. Dispatch is a registered trademark of Caltech Industries, Inc. Vacutainer is a registered trademark of Becton, Dickinson and Company. All other product names and trademarks are property of their respective owners. 2007-2013 Roche Diagnostics. All rights...»

«PREFACE TO SELECTED INFORMATION DIRECTIVES   CIO Transmittal No.: 15-010 CIO Approval Date: 06/12/2015 Issued by the EPA Chief Information Officer, Pursuant to Delegation 1-19, dated 07/07/2005 CHIEF INFORMATION OFFICER MEMORANDUM   SUBJECT: Chief Technology Officer (CTO) Responsibilities in Selected Information Directives Re-assigned CTO responsibilities Effective immediately, CTO responsibilities detailed in the selected information directives (i.e., Information Policies, Procedures,...»

«Read & Download The Game That Breaks Us Online Library ~ PDF Book For FREE Library The Game That Breaks Us Download | Read Full Book Total Downloads: 44169 Formats: AZW | ODF | Audio | ePub | Mobi | LIT | Kindle | PDF Rated: 8/10 (3712 votes) The Game That Breaks Us Hockey’s bad boy needs to clean up his reputation. Bennett James has built his career at being the best, on and off the ice, and now it’s come back to bite him. The media has turned against him, and with a near career-ending...»

«by Samantha Ip Clark Wilson LLP tel. 604.643.3172 ssi@cwilson.com D/RJR/682791.1 TABLE OF CONTENTS A. INTRODUCTION B. GENERAL INFORMATION ON THE TYPES OF LIABILITY INSURANCE 1. GENERAL LIABILITY INSURANCE POLICIES Wrap-Up Insurance Policies 2. PROFESSIONAL LIABILITY INSURANCE POLICIES ClaimsMade vs. Occurrence Basis C. TYPES OF PROFESSIONAL LIABILITY POLICIES 1. PRACTICE INSURANCE 2. PROJECT INSURANCE D. PRACTICAL ADVICE WHEN OBTAINING CONSULTANTS’ LIABILITY INSURANCE.9 E. CONCLUSION Samantha...»

«MINUTES MONTANA HOUSE OF REPRESENTATIVES 51st LEGISLATURE REGULAR SESSION COMMITTEE ON STATE ADMINISTRATION Call to Order: By Chairman Jan Brown, on January 24, 1989, at 9:00 a.m. ROLL CALL Members Present: All present except: Members Excused: Rep. Janet Moore Members Absent: None Staff Present: Judy Burggraff, Secretary; Lois Menzies, Staff Researcher Announcements/Discussion: Subcommitt~e Chairman Cocchiarella said that she had just received a fiscal note that morning for HB 26, and the...»

«Publicada D.O. 19 set/005 Nº 26835 Ley Nº 17.897 LIBERTAD PROVISIONAL Y ANTICIPADA SE ESTABLECE UN RÉGIMEN EXCEPCIONAL El Senado y la Cámara de Representantes de la República Oriental del Uruguay, reunidos en Asamblea General, DECRETAN: CAPÍTULO I DEL RÉGIMEN EXCEPCIONAL DE LIBERTAD PROVISIONAL Y ANTICIPADA Artículo 1º. (Libertad anticipada y provisional excepcionales).El régimen excepcional de libertad anticipada y provisional que se establece en la presente ley se aplicará, por...»

«City of Pomona The State of Homelessness in Pomona Overview, Analysis and Recommendations Prepared By Mark Lazzaretto, CDD Director Benita DeFrank, Housing Manager Jan Cicco, Homeless Services Coordinator Table of Contents FORWARD INTRODUCTION HISTORICAL RESPONSE TO HOMELESSNESS IN POMONA GUIDING PRINCIPLES OF HOMELESS SERVICES IN POMONA HOW IS HOMELESSNESS DEFINED IN POMONA? A 10-YEAR OVERVIEW OF HOMELESS SERVICES IN POMONA PLANNING TOOLS FOR A STRATEGY BI-ANNUAL COUNT 2013 HOMELESS CONSUMER...»

«BWP BRL 13 COP COU EUR IDR PEN NL,SGDJMD,ANG, THB, PAB, ETB,MKD,MT WL, VND, GRD, AMD, XDR, CVE, MZE, PTE NOK, SKK, SEK, CZK, CSK, GMD, SZL, DZ PE, AOK,MDL, ROL,KWD, BGL,GEL, NZD, ADF FA,SLL, UAH, PGK,RON, LYD, BGN,LVL, SGD A, SML,IQD, JOD, CYP,AWG,FIM,MWK, ZMK, G, TMM,VAL, LTL,BBD, MMK,NAD, HUF, MZN KD,EUR, AON, MGA, LRD, FKP, GIP,VEB,LBP AFN, CHE, AOR, NLG, PYG, TMT, DEM,LAK, HRK,RSD, CSD,VEF SBD,XAG, L, ZARCRC, SVC, BAM, USN,MZM, MGF BHD, SDG, SYP, TRL, TRY, GBP, LSL, ALLSD EGP, GNF, AUD, BSD...»

«PERCEPTIONS OF EMPOWERMENT BY THE CONSTRUCTION SENIOR MANAGEMENT Kay Greasley1, Nicola King1 Alan Bryman2, Andrew R.J. Dainty1 and Andrew D.F. Price1 Department of Civil and Building Engineering, Loughborough University, Leicestershire, UK Department of Social Sciences, Loughborough University, Leicestershire, UK The research reviews the potential of team and individual empowerment as performance improvement strategies for construction. The important role of empowerment has been emphasized...»

«An explanation of how to apply the unique powders of the Creation Porcelain System THE CREATION KIT OPACIOUS DENTINS THE CREATION KIT contains Opacious Dentins for the sixteen (16) Vita shades A-1 to D-4. Opacious Dentins were developed by Willi Geller to help control areas that are notably dark and shadowed. Opacious Dentins are used to control both the value and the chroma of a crown, and are inherently more opacious to help reduce the amount of light transmitted in the mouth. When too much...»

«IN THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF PENNSYLVANIA MICHAEL TOMLINSON and SHERRY : CIVIL ACTION TOMLINSON, : : Plaintiffs, : NO. 09-3711 v. : :BOROUGH OF NORRISTOWN, POLICE : OFFICER LORI KOCH, and POLICE : OFFICER BEN LERMAN : : Defendants. : MEMORANDUM ON MOTION FOR PARTIAL DISMISSAL Baylson, J. April 5, 2010 Plaintiffs Michael and Sherry Tomlinson bring several civil rights claims against Defendants the Borough of Norristown (“Borough”), Police Officer Lori Koch,...»

«Christmas Carols in British Sign Language Mansfield Deaf Club 6 1 Wood Street, Mansfield, Notts NG18 1QB, 3.30pm. Nottinghamshire Deaf Society 16 22 Forest Road West, Nottingham NG7 4EQ, 8pm. Attenborough St Mary’s 6 Toys & Gifts & Christingle Service (please bring new toys and gifts for children in need in Notts; collection for Children’s Society), 4.30pm. 13 Service of Nine Lessons & Carols, 6.30pm. 20 Family Carol Service, 5.30pm. Christmas Eve Crib Service (children are welcome to come...»





 
<<  HOME   |    CONTACTS
2016 www.theses.xlibx.info - Theses, dissertations, documentation

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.