In the year of the restructuring, fewer teachers will be retained in their same position than in other periods.The option value of retaining an inexperienced teacher with low posterior mean but high variance is higher than for an experienced teacher with the same posterior mean (so better average performance to date) but low varianceŠthe inexperienced teacher may turn out to be fine, and can always be fired next year if she doesn™t. In the year of the layoffs, there is an inverse relationship between an untenured teacher's experience and probability of returning to his or her same position.In the year of the layoffs, there is a positive relationship between probability of retention and teacher performance as measured by value-added score. 1.6.1 Change in Retention Rates 1.6.2 Characteristics of Teachers Retained 1.6.3 Impact of Layoffs on Student Test Scores 1.6.4 Synthetic Control Model Robustness Check 1.6.5 Additional Placebo and Robustness Tests 2.6.1 Effect of Layoff Announcement on Teacher Absences 2.6.2 Effect of Layoff Announcement on Teacher Absences by Month, Week, and Day 2.6.3 Effect of Layoff Announcement on Teacher Absences by Subject Taught, Experience Level, and Value-Added Tercile 2.6.4 Testing the Principal Tip-Off Hypothesis 2.6.5 Placebo and Robustness Tests of Main Results existingnegativereduce i st Charter.3.7.1 Effect of Charter Penetration on Housing Prices 3.7.2 Testing for Endogenous Charter Location 3.7.3 Effect of Charter Penetration on Housing Prices: Heterogeneity and Specification Checks Notes: The first listed year refers to the fiscal year used to calculate the initial number of teachers. The second listed year refers to the fiscal year that we observe whether a teacher returns to his/her same position or not. The highlighted rows indicate the year of the mass teacher layoffs. The calculation for the retention variable is if a teacher was ever seen returning to the same school. School-clustered standard errors are reported in parentheses. Teacher-clustered standard errors reported in brackets.Dummy for Teacher Returning to Same SchoolUntenured Teacher in RPS in 2010 Untenured Teacher in RPS First-Year Teacher in RPS in 2010 Second-Year Teacher in RPS in 2010 Third-Year Teacher in RPS in 2010 Fourth-Year Teacher in RPS in 2010 First-Year Teacher in RPS Second-Year Teacher in RPS Third-Year Teacher in RPS Fourth-Year Teacher in RPS Year Fixed Effects District Fixed Effects Main Effects Included Total Number of Observations Number of Teachers Number of Untenured Teachers Notes: School-clustered standard errors are in parentheses. Marginal effects for Logit regressions are reported. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Same School Left District Switched Schools Untenured RPS Teacher in 2010 Untenured RPS Teacher Year Fixed Effects Total Number of Observations Number of Teachers Number of Untenured Teachers Notes: School-clustered standard errors are in parentheses. In the main regression, Same Position is the omitted category and then is backed out from the results. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Dummy for Teacher Returning to Same SchoolUntenured Teacher Untenured Teacher in 2010 Dummy for Teacher Leaving the DistrictUntenured Teacher Untenured Teacher in 2010 Year Fixed Effects Dummy Control Variables for Teacher Experience Controls for Main Effects Total Number of Observations Number of Teachers Number of Untenured Teachers Notes: School-clustered standard errors are in parentheses. Teacher observable variables are included in the regressions but not reported. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Reading Z-ScoreMath Z-ScorePercent of Teachers Untenured in 2010*Year After Layoffs Percent of Teachers Untenured in 2010*2 Years After Layoffs Year Fixed Effects School Fixed Effects Control for Main Effects Student Fixed Effects Lag of Student Test Score Total Number of Observations Number of Individual Students Notes: School-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Percent of Teachers Leaving the School in 2010 (interacted with year dummies)Percent of Teachers Untenured in 2010*Year After Layoffs Percent of Teachers Untenured in 2010*2 Years After Layoffs Reading Z-ScoreMath Z-ScorePercent of Teachers Leaving the School in 2010*Year After Layoffs Percent of Teachers Leaving the School in 2010*2 Years After Layoffs Year Fixed Effects School Fixed Effects Control for Main Effects Student Fixed Effects Lag of Student Test Score Total Number of Observations Number of Individual Students Notes: Student-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Dummy for Teacher Returning to Same SchoolUntenured Teacher in RPS in 2010 Untenured Teacher in RPS First-Year Teacher in RPS in 2010 Second-Year Teacher in RPS in 2010 Third-Year Teacher in RPS in 2010 Fourth-Year Teacher in RPS in 2010 First-Year Teacher in RPS Second-Year Teacher in RPS Third-Year Teacher in RPS Fourth-Year Teacher in RPS Year Fixed Effects Total Number of Observations Number of Teachers Number of Untenured Teachers Notes: For Synthetic control, teacher-clustered standard errors are in parentheses. For main results, school-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Dummy for Teacher Returning to Same SchoolUntenured Teacher in RPS in 2010 First-Year Teacher in RPS in 2010 Second-Year Teacher in RPS in 2010 Third-Year Teacher in RPS in 2010 Fourth-Year Teacher in RPS in 2010 Year Fixed Effects District Fixed Effects Main Effects Included Total Number of Observations Number of Teachers Number of Untenured Teachers Notes: School-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Dummy for Teacher Returning to Same SchoolUntenured Teacher in RPS in 2009 Untenured Teacher in RPS in 2008 Year Fixed Effects District Fixed Effects Main Effects Included Total Number of Observations Number of Teachers Number of Untenured Teachers Notes: School-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Dummy for Teacher Returning to Same SchoolUntenured Teacher in RPS in 2008 Untenured Teacher in RPS in 2009 Untenured Teacher in RPS in 2010 Untenured Teacher in RPS in 2011 Year Fixed Effects Total Number of Observations Number of Individual Teachers Number of Untenured Teachers Notes: Teacher-clustered standard errors are in parentheses. Marginal effects for Logit regressions are reported. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Percent of Teachers Untenured in 2008 Percent of Teachers Untenured in 2008*One Year After Year Fixed Effects School Fixed Effects Student Fixed Effects Total Number of Observations Number of Individual Students Notes: School-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Percent of Male StudentsPercent of Black StudentsPercent of Hispanic StudentsPercent of Low-Income StudentsAverage Lagged Reading Z- ScoreAverageLagged Math Z- ScorePercent of Untenured Percent of Untenured Teachers in 2010*Year of the Layoffs Percent of Untenured Teachers in 2010*Year After Layoffs Percent of Untenured Teachers in 2010*2 Years After Layoffs Year Fixed Effects School Fixed Effects Total Number of Observations Number of Individual Schools Notes: School-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Panel A: Average Percent of School Days Absent During Layoff PeriodUntenured Tenured Young Tenured Panel B: Average Number of School Days Absent During Layoff PeriodUntenured Tenured Young Tenured Notes: Teacher-clustered standard errors are reported in parentheses.The layoff period is the time between the layoffs were announced (March 23rd) and the end of the school year (June 30). Panel A - OLS - Percent of Days Absent Over Layoff Period Untenured Teacher*Layoffs Announced Panel B - OLS - Number of Days Absent Over Layoff Period Untenured Teacher*Layoffs Announced Panel C - PQMLE - Number of Days Absent Untenured Teacher*Layoffs Announced Year Fixed Effects Main Effects Included Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. The layoff period is the time between the layoffs were announced (March 23rd) and the end of the school year (June 30).PQMLE regressions all include teacher fixed effects and have fewer observations. The total number of observations for the PQMLE regressions is 658 and the number of individual teachers is 316. Number of Days Absent by MonthUntenured*March 2010 Untenured*April 2010 Untenured*May 2010 Month and Year Fixed Effects Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Number of Days AbsentUntenured*Week of Layoff Announcement Untenured*Week 1 of April 2010 Untenured*Week 2 of April 2010 Untenured*Week 3 of April 2010 Untenured*Week 4 of April 2010 Untenured*Week 1 of May 2010 Untenured*Week 2 of May 2010 Untenured*Week 3 of May 2010 Untenured*Week 4 of May 2010 Week and Year Fixed Effects Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Number of Days AbsentUntenured*March 23 Untenured*March 24 Untenured*March 25 Untenured*March 26 Untenured*March 29 Untenured*March 30 Untenured*March 31 Untenured*April 1 Day and Year Fixed Effects Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Number of Days AbsentBottom School Tercile Untenured Teacher in 2010 Second School Tercile Untenured Teacher in 2010 Top School Tercile Untenured Teacher in 2010 Annual Fixed Effects School Fixed Effects Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. No constant is included in these regressions. Number of School Days AbsentUntenured Math Teacher*Layoff Period Untenured Science Teacher*Layoff Period Untenured Reading Teacher*Layoff Period Untenured Self-Contained Elementary Teacher*Layoff Period Untenured Other Subject Teacher*Layoff Period Annual Fixed Effects Main Effects Included Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Number of Days AbsentFirst-Year Teacher*Layoffs Announced Second-Year Teacher*Layoffs Announced Third-Year Teacher*Layoffs Announced Fourth-Year Teacher*Layoffs Announced Annual Fixed Effects Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Number of Days AbsentReading VAM Tercile 1 Untenured Teacher in Year of Layoffs Reading VAM Tercile 2 Untenured Teacher in Year of Layoffs Reading VAM Tercile 3 Untenured Teacher in Year of Layoffs Math VAM Tercile 1 Untenured Teacher in Year of Layoffs Math VAM Tercile 2 Untenured Teacher in Year of Layoffs Math VAM Tercile 3 Untenured Teacher in Year of Layoffs Annual Fixed Effects Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. No constant is included in these regressions. Number of Days AbsentPanel A - Ordinary Least Squares Untenured Teacher*Layoffs Announced Untenured Teacher in Same Position the Following Year*Layoffs Panel B - Poisson Quasi-Maximum Likelihood Estimation Untenured Teacher*Layoffs Announced Untenured Teacher in Same Position the Following Year*Layoffs Annual Fixed Effects Main Effects Included Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Number of Days Absent Untenured*Week of Layoff Announcement Untenured*Week 1 of April 2010 Untenured*Week 2 of April 2010 Untenured*Week 3 of April 2010 Untenured*Week 4 of April 2010 Untenured*Week 1 of May 2010 Untenured*Week 2 of May 2010 Untenured*Week 3 of May 2010 Untenured*Week 4 of May 2010 Week and Year Fixed Effects Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Number of Days AbsentUntenured*March 16 Untenured*March 17 Untenured*March 18 Untenured*March 19 Untenured*March 20 Untenured*March 23 Untenured*March 24 Untenured*March 25 Untenured*March 26 Untenured*March 29 Untenured*March 30 Untenured*March 31 Untenured*April 1 Day and Year Fixed Effects Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Control for Number of School Days Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Conversion charters Start-up charters Property Characteristics Number of Charters Charters as percentage of enrollment A: Characteristics of zoned school B: Characteristics of charters within 1 mile (enrollment weighted) A. Distance gradient B. Condensed 0-2 miles Start-up charters Conversion charters B. Condensed 0-2 miles Reading Math Single-Lag VAM Double-Lag VAMSingle-Lag VAMDouble-Lag VAM Dummy for Teacher Returning to Same SchoolFirst Year Superintendant Year Fixed Effects Total Number of Observations Number of Untenured Teachers Notes: Teacher-clustered standard errors are in parentheses. Marginal effects for Logit regressions are reported. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Dummy for Teacher Being Rehired in Same PositionUntenured Female Teacher in RPS in 2010 Untenured White Teacher in RPS in 2010 Untenured Teacher Log Salary in RPS in 2010 Year Fixed Effects District Fixed Effects Main Effects Included Total Number of Observations Number of Teachers Number of Untenured Teachers Notes: Teacher-clustered standard errors are in parentheses. Marginal effects for Logit regressions are reported. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Same School Left District Switched Schools First-Year Teacher in RPS in 2010 Second-Year Teacher in RPS in 2010 Third-Year Teacher in RPS in 2010 Fourth-Year Teacher in RPS in 2010 First-Year Teacher in RPS Second-Year Teacher in RPS Third-Year Teacher in RPS Fourth-Year Teacher in RPS Year Fixed Effects Total Number of Observations Number of Teachers Number of Untenured Teachers Notes: School-clustered standard errors are in parentheses. In the main regression, Same School is the omitted category and then is backed out from the results. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Untenured Leavers in 2010 New School in 2011 New to RPS in 2011 ƒ The sample sizes for the subsets of teachers with value-added scores are Leavers = 17, New School = 44, and New to RPS = 18. Byron 25.60% City of Chicago 20.80% Galesburg 0.80% North Boone 1.80% Plano 0.90% Triad 25.80% Wauconda 1.40% Waukegan 1.10% Other Illinois Districts (each between 0.1 and 0.6%) 21.80% Total 100.00% Number of Days Absent Over Layoff Period Panel A - OLS - Percent of Days Absent Over Layoff Period Untenured Teacher*Layoffs Announced Untenured Teacher*Layoffs Rumored Panel B - OLS - Number of Days Absent Over Layoff Period Untenured Teacher*Layoffs Announced Untenured Teacher*Layoffs Rumored Panel C - PQMLE - Number of Days Absent Untenured Teacher*Layoffs Announced Untenured Teacher*Layoffs Rumored Annual Fixed Effects Main Effects Included Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. The layoff period is the time between the layoffs were announced (March 23rd) and the end of the school year (June 30). PQMLE regressions all include teacher fixed effects and have fewer observations. The total number of observations for the PQMLE regressions is 4,616 and the number of individual teachers is 1,420. Number of Days AbsentPanel A - Main Results - District Experience Definition of Tenure Untenured Teacher*Layoffs Announced Panel B - State Experience Definition of Tenure Untenured Teacher*Layoffs Announced Panel C - Total Experience Definition of Tenure Untenured Teacher*Layoffs Announced Annual Fixed Effects Main Effects Included Teacher Fixed Effects Linear Control Variable for Experience Dummy Control Variables for Experience Total Number of Observations Number of Individual Teachers Notes: Teacher-clustered standard errors are in parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level. Elementary - high school Middle school High school Elementary school A. Number of charters B. Charter seats as percentage of enrollment A. Distance gradient Housing Characteristics Local School Characteristics B. Condensed 0-2 miles Quarterly Journal of Economics Journal of Policy Analysis and Management NBER Working Paper #17332 Working Paper Journal of Political Economy Economics of Education Review Education Finance and Policy Quarterly Journal of Economics Handbook of the Economics of Education, Volume 3 National Tax Journal Journal of Urban Economics Journal of Urban EconomicsJournal of Public Economics Journal of Policy Analysis and Management The Quarterly Journal of Economics Federal Reserve Bank of New York Staff Report #472 Journal of Urban Economics American Economic Review The National Study of Charter Management Organization (CMO) Effectiveness. Journal of Urban Economics Journal of Urban Economics Journal of Public Economics Journal of Public Economics . The Review of Economics and Statistics, Journal of Urban Economics. American Law and Economic Review American Educational Research Journal The Economics of School Choice Journal of Political Economy Journal of Urban Economics Journal of Political Economy Education Finance and Policy National Tax Journal Journal of Urban Economics Public Administration Review