The sample contains the pooled cross-section of students living within 3 miles of a killing from expected grades 9 through 16, where expected grade is determined by the year students began 9th grade at LAUSD. Author Guidelines; Submission Site; Open Access; Purchase; About. As my preferred estimating equation includes census block groupâsemester fixed effects, causal identification comes from comparing treatment and control students in the same census block group. Second, this article complements a growing body of research demonstrating how perceived discrimination may lead to âself-fulfilling propheciesâ in education (Carlana 2019), labor markets (Glover, Pallais, and Pariente 2017), and health care (Alsan and Wanamaker 2018).7 While empirical evidence of racial bias is mixed (Nix etÂ al. Regardless of demographics or circumstance, involved officers were rarely prosecuted. The Online Appendix also provides evidence that the long-run effects are not driven by differential attrition (i.e., students transferring out of LAUSD).47 In particular, Online Appendix Figure A.XIII decomposes the effect on high school graduation by estimating equation (4) on an indicator for whether a student transferred out of LAUSD and, separately, on an indicator for whether a student dropped out altogether (i.e., did not graduate and did not transfer). Newspaper mentions come from a search of each incident by the name of the deceased in six local newspapers (the Los Angeles Times, the Los Angeles Daily News, Pasadena Star News, San Gabriel Valley Tribune, Torrance Daily Breeze, and Whittier Daily News). I recover larger treatment estimates for male students as well as for students with less-educated parents or lower 8th-grade test scores, suggesting that lower-achieving and more disadvantaged students may be most affected by exposure to police killings. In particular, I control for whether the incident occurred near the deceasedâs home and whether it was mentioned in a local newspaper, as students may be more affected by killings that involved someone they personally knew or that were more visible.37 Finally, I control for age and gender of the deceased in column (4) to account for the fact that blacks and Hispanics killed by police were younger on average than whites and Asians. The effects are driven entirely by black and Hispanic students in response to police killings of other underrepresented minorities. Taken together, the results indicate that police killings may have large long-run effects on local communities. Roughly 20% of incidents involved unarmed people, approximately the same share as those that involved armed individuals who fired at others. Schultz, Dayan, and Montague (1997) noted that the shift forward in time of dopamine release in response to information about a future reward sug-gests that it plays a role in reinforcement learning (Bush and Mosteller 1951; Rescorla and Wagner 1972). Comparing across the four specifications, results mirror those found in Figure IV with significant, negative treatment effects for black/Hispanic killings of around 0.030 points and insignificant, near-zero estimates for white/Asian killings that never rise above 0.008 points in magnitude. An Online Appendix for this article can be found at The Quarterly Journal of Economics online. To account for this, Shootâ7 and Shoot7 are set to 1 in cases where Ï < â7 and Ï > 7, respectively. From the 2004 school year onward, I observe the date students were designated by the district as âemotionally disturbed,â a federally certified learning disability that âcannot be explained by intellectual, sensory or health factorsâ and that qualifies for special education accommodations. QUARTERLY JOURNAL OF ECONOMICS Vol. Even if measurement error was correlated with treatment, the inclusion of student fixed effects would account for any level biases that might arise due to migrationâsuch as if high-achieving students were more likely to relocate following exposure.30. CXX August 2005 Issue 3 DIVIDEND TAXES AND CORPORATE BEHAVIOR: EVIDENCE FROM THE 2003 DIVIDEND TAX CUT* RAJ CHETTY AND EMMANUEL SAEZ This paper analyzes the effects of dividend taxation on corporate behavior QUARTERLY JOURNAL OF ECONOMICS Vol. Instead, my findings suggest a more nuanced story about race-match: conditional on exposure, black and Hispanic students respond differently to police violence depending on the race of the person killed. Column (5) instead controls for neighborhood-cohort using arbitrary square mile units derived from dividing Los Angeles into a grid. The Quarterly Journal of Economics, 112, 1203-1250. In particular, I estimate a simplified model version of equation (3) regressing semester GPA on the number of police and criminal killings of whites/Asians and blacks/Hispanics a student was exposed to in the current and prior semester. 2001). 2018 : 11/25/19 : Cambridge Journal of Economics : Accepted : 5 : 5 : 1 : Professional co-editor and referee. Only 10% of individuals were of school age (i.e., 19 or younger), and none were actively enrolled LAUSD students. eighth-grade CST score (by pentile), cohort (within three years), and school. Column (6) expands the sample to allow students to appear as both treatment and control in a given expected grade (i.e., if the student lived within 0.50 miles of a killing and between 0.50 and 3 miles of a different killing in that grade). Consistent with research by Derenoncourt (2018) documenting a negative correlation between police presence and black upward mobility in Great Migration destinations, my results suggest that law enforcement may play an important role in explaining this racial disparity. The sample size is the last column. At the same time, the four largest urban riots in recent U.S. history were all triggered by acts of police violence (DiPasquale and Glaeser 1998).1 Experiences with aggressive policing have been linked to unfavorable attitudes toward law enforcement, particularly among racial minorities, whose lifetime odds of being killed by police are as high as 1 in 1,000 (Skolnick and Fyfe 1993; Weitzer and Tuch 2004; Brunson and Miller 2005).2 These attitudes are, in turn, correlated with fear (Hale 1996; Renauer 2007; Boyd 2018), perceived discrimination (Brunson 2007; Carr, Napolitano, and Keating 2007), and institutional distrust (Bobo and Thompson 2006; Kirk and Papachristos 2011). Using cross-sectional data, the first difference in a DD model would compare graduation rates of students in expected grades â¤12 living nearby (within 0.50 miles) to graduation rates of nearby students in expected grades >12, where expected grade is determined by the year a student began 9th grade at LAUSD. The ninth-grade treatment estimates correspond to a 3.4% decrease in graduation rate (mean of 50%) and a 2.7% decrease in postsecondary enrollment rate (mean of 32.6%). 2 QUARTERLY JOURNAL OF ECONOMICS necessary permits. (2020) find that intergenerational mobility differs dramatically between blacks and whites, even for children from the same neighborhood and socioeconomic background. The sample includes student-semester panel data from the 2010 calendar year onward, the period for which crime data are available. However, as reflected by nationwide protests over the police killings of Michael Brown and George Floyd, community perceptions of âreasonablenessâ often depend on contextual factors similar to those assessed here, with police violence against unarmed minorities drawing particular concern (Hall, Hall, and Perry 2016). Point estimates for reported crimes never exceed 0.31 in magnitude, less than 10% of the sample mean (3.16 reported crimes per block-semester). Advocates of the EITC argue that the credit transfers income to a particularly deserving group of people, the working poor, and that the redistribution occurs with much less distortion of labor supply than is caused by other elements of the welfare system. Area characteristics include population density, average income, homicide rate, and percent nonwhite in a studentâs block group. Panel A depicts the absenteeism gradient of distance, separately for the week before police killings and the week after (including the incident date). The latter is an outcome of interest in its own right, which I examine in Section VI. By geocoding the exact location of the 627 incidents and over 700,000 home addresses, I can calculate each studentâs precise geographic proximity to police violence. 2 THE QUARTERLY JOURNAL OF ECONOMICS I. Burdick-Will Julia, Ludwig Jens, Raudenbush Stephen W., Sampson Robert J., Sanbonmatsu Lisa, Sharkey Patrick. Sample includes student-semester panel data for students enrolled in LAUSD high schools from the 2002â2003 academic year to the 2015â2016 academic year. In contrast, an entrepreneur in Canada can finish the process in two days by paying US\$280 in fees and completing only two procedures. While the survey is not mandatory, it is typically administered during school hours leading to response rates above 75%. Panel A provides summary statistics for the full police killings data and separately for killings of blacks and Hispanics and killings of whites and Asians. Standard errors are clustered by ZIP code. See Butter” eld [1997a, ©2001 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology. The tools utilized for analyzing the problem comprise, a coefficient of protection analysis as well as a supply response analysis by means of a time series regression (for the period 1980-2011). Semester GPA is calculated from student transcript data. The area bore witness to two of the most high-profile acts of police violence in U.S. history. As shown, I find stark differences in effects by student race. Evidence from the Great Migration,â, âPolice Violence and Citizen Crime Reporting in the Black Community,â, âThe Los Angeles Riot and the Economics of Urban Unrest,â, âRisk of Being Killed by Police Use of Force in the United States by Age, RaceâEthnicity, and Sex,â, Proceedings of the National Academy of Sciences, âPersistence of Pediatric Post Traumatic Stress Disorder after 2 Years,â, Local Polynomial Modelling and Its Applications, âPrevalence of Serious Emotional Disturbance in Children and Adolescents,â, âThe Production of Human Capital in Developed Countries: Evidence from 196 Randomized Field Experiments,â, âHandbook of Economic Field Experimentsâ, âAn Empirical Analysis of Racial Differences in Police Use of Force,â, âPolice Shootings, Civic Unrest and Student Achievement: Evidence from Ferguson,â, âThe Effect of Community Traumatic Events on Student Achievement: Evidence from the Beltway Sniper Attacks,â, âDiscrimination as a Self-fulfilling Prophecy: Evidence from French Grocery Stores,â, âDifference-in-Differences with Variation in Treatment Timing,â, âDevelopmental Differences in Risk Factors for Suicide Attempts between Ninth and Eleventh Graders,â, âFear of Crime: A Review of the Literature,â, âBlack and Blue: Exploring Racial Bias and Law Enforcement in the Killings of Unarmed Black Male Civilians,â, âRole of Trauma in the Lives of High School Dropouts,â, âTesting, Stress, and Performance: How Students Respond Physiologically to High-Stakes Testing,â, âViolence and Vigilance: The Acute Effects of Community Violent Crime on Sleep and Cortisol,â, âHow Kids View Cops the Nature of Juvenile Attitudes toward the Police,â, âThe Determinants of Deadly Force: A Structural Analysis of Police Violence,â, âOfficer Characteristics and Racial Disparities in Fatal Officer-Involved Shootings,â, âPolice Violence as a Function of Community Characteristics,â, âMoving to Opportunity in Boston: Early Results of a Randomized Mobility Experiment,â, Report of the National Advisory Commission on Civil Disorders, âCultural Mechanisms and the Persistence of Neighborhood Violence,â, âAdministrative Records Mask Racially Biased Policing,â, âMaking Inferences about Racial Disparities in Police Violence,â, âExploring the Policy Implications of High-Profile Police Violence,â, âAggressive Policing and the Educational Performance of Minority Youth,â, âExplaining Juvenilesâ Attitudes toward the Police,â, âThe Effect of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports,â, âVulnerability and Resilience: A Study of High-Risk Adolescents,â, âDifferences between Gang and Nongang Homicides,â, âServices to Children with Serious Emotional Disturbance,â, âDrug Battles and School Achievement: Evidence from Rio De Janeiroâs Favelas,â, âViolence, Psychological Trauma, and Risk Attitudes: Evidence from Victims of Violence in Colombia,â, âTrauma and PTSD among Adolescents with Severe Emotional Disorders Involved in Multiple Service Systems,â, âA Birdâs Eye View of Civilians Killed by Police in 2015: Further Evidence of Implicit Bias,â, âOn Views of Race and Inequality, Blacks and Whites Are Worlds Apart,â, âChildhood Trauma and Psychiatric Disorders as Correlates of School Dropout in a National Sample of Young Adults,â, âAn Honest Approach To Parallel Trends,â, âReducing Fear of Crime: Citizen, Police, or Government Responsibility?â, âLocal Exposure to School Shootings and Youth Antidepressant Use,â, âThe Acute Effect of Local Homicides on Childrenâs Cognitive Performance,â, âHigh Stakes in the Classroom, High Stakes on the Street: The Effects of Community Violence on Studentâs Standardized Test Performance,â, âThe Effect of Local Violence on Childrenâs Attention and Impulse Control,â, âPolice Brutality and Public Perceptions of Racial Discrimination: A Tale of Two Beatings,â, Above the Law: Police and the Excessive Use of Force, Juvenile Offenders and Victims: 1996 Update on Violence, US Department of Justice, Office of Justice Programs, âEstimating Dynamic Treatment Effects in Event Studies With Heterogeneous Treatment Effects,â, âThe County: The Story of Americaâs Deadliest Police,â, âCoppinâ an Attitude: Attitudinal Differences among Juveniles toward Police,â, âStreet Stops and Police Legitimacy: Teachable Moments in Young Urban Menâs Legal Socialization,â, âTo Assure the Free Appropriate Public Education of All Children with Disabilities: Fifteenth Annual Report to Congress on the Implementation of the Individuals with Disabilities Education Act,â, âOutcomes for Youths with Serious Emotional Disturbance in Secondary School and Early Adulthood,â, âPerceptions of Racial Profiling: Race, Class, and Personal Experience,â, âRace and Perceptions of Police Misconduct,â, âExamining the Impact of the Freddie Gray Unrest on Perceptions of the Police,â, âWhat Weâre Missing: A Descriptive Analysis of Part-Day Absenteeism in Secondary School,â, âAdolescent Attitudes towards the Police: A Survey of High School Students,â, Juvenile Delinquency: Little Brothers Grow Up, âEffects of Trauma Exposure on the Cortisol Response to Dexamethasone Administration in PTSD and Major Depressive Disorder,â. Greater interrogation of these concerns and a more nuanced understanding of the specific factors driving them is vital to determining the optimal levels and responsibilities of law enforcement. Chetty etÂ al. Nonetheless, information about weapon type and discharge has the benefit of being objectively verifiable and can be found in all available incident reports. In other cases, killings were precipitated by seemingly innocuous encounters that escalated unexpectedly. : Rising Inequality, Schools, and Children's Life Chances, (New York: Russell Sage Foundation, 2011), 255â276, âDifference-in-Differences with Multiple Time Periods,â, âViolence and Risk Preference: Experimental Evidence from Afghanistan,â, âRobust Inference with Multiway Clustering,â, Journal of Business & Economic Statistics, âImplicit Stereotypes: Evidence from Teachersâ Gender Bias,â, âWe Never Call the Cops and Here Is Why: A Qualitative Examination of Legal Cynicism in Three Philadelphia Neighborhoods,â, âExternalities in the Classroom: How Children Exposed to Domestic Violence Affect Everyoneâs Kids,â, American Economic Journal: Applied Economics, âSelf-determination Skills and Opportunities of Transition-Age Youth with Emotional Disturbance and Learning Disabilities,â, âThe Value of School Facility Investments: Evidence from a Dynamic Regression Discontinuity Design,â, âAre US Cities Underpoliced? Eighty-six percent of students identify as either black or Hispanic, while only 14% are white or Asian.18 The majority of students come from disadvantaged households, with 69% qualifying for free or subsidized lunch and less than 10% with college-educated parents. Columns (4) and (5) replace the cohort by census block group fixed effects with cohort by census tract and cohort by square-mile grid units, respectively. However, as I discuss in Section V, effect sizes do increase with the demographic similarity of students and deceased, suggesting that informal networks or personal affiliation may be a more salient mediating channel. However, across specification, the weapon gradient becomes steeper when restricting to killings of blacks and Hispanics. This form allows you to upload a manuscript, cover letter, or other files or supporting documentation for your submission (or resubmission) to Quarterly Journal of Economics.. However, because I am able to account for time trends at the neighborhood level, any potential biases would have to be hyperlocal, differentially affecting students in the same census block group. Instead, I find that the most damaging events are police killings of unarmed individuals, those who may have been the least likely to pose a threat to the community or to be engaged in a violent crime at the time of the incident. The Quarterly Journal of Economics is the oldest professional journal of economics in the English language. Column (2) restricts the treatment group to students living within 0.25 miles of a killing in an expected grade. Furthermore, most existing studies examine effects on attitudes or interactions with law enforcement and are unable to shed light on broader economic implications. Download online appendix. These effects are driven entirely by black and Hispanic students in response to police killings of other minorities and are largest for incidents involving unarmed individuals. Instead of the continuous limit order book market design that is currently Over a short period of time in the early 1970s, that pattern changed 2017; Fryer 2019; Johnson etÂ al. I searched for each incident by the name of the deceased in six local newspapers (the Los Angeles Times, the Los Angeles Daily News, Pasadena Star News, San Gabriel Valley Tribune, Torrance Daily Breeze, and Whittier Daily News). 434 QUARTERLY JOURNAL OF ECONOMICS not hold. Additional information about the data is included in the Online Appendix. Specifically, I estimate local polynomial regressions of daily absenteeism on the distance between a studentâs home and the incident location. Correlational analysis of police violence and neighborhood health is confounded by the fact that use of force is more likely to occur in disadvantaged areas, where homicide and poverty rates are high (Kania and Mackey 1977; Jacobs 1998). Each report includes a detailed description of the event based on forensic and investigative evidence and officer and witness testimonies. \end{eqnarray}$$. However, duplicates are removed such that a given student may only appear once per expected grade. These include the 1965 Watts riots, the 1980 Miami riots, the 1992 Los Angeles riots, and the 2013 Ferguson riots. Includes demographic controls for a studentâs school, race, sex, free lunch status, household language, parental education, and eighth-grade proficiency. Callaway Brantly, SantâAnna Pedro H. C.. Callen Michael, Isaqzadeh Mohammad, Long James D., Sprenger Charles. 2011; Burdick-Will 2013; Sharkey etÂ al. Table A.I shows results using alternative calculations of standard errors (i.e., multiway clustering with ZIP code and year and clustering by school catchment or tract). Please check the relevant section in this Guide for Authors for more details. As corroboration, I find that exposed students are 15% more likely to be classified with emotional disturbanceâa chronic learning disability associated with PTSD and depressionâand twice as likely to report feeling unsafe in their neighborhoods the following year. All appendix tables and figures mentioned in the text are included in the Online Appendix. Consistent with recent work demonstrating that violence affects cortisol levels (Heissel etÂ al. This is done by estimating the following event study model on the pooled data:$$ This is consistent with the exogeneity of police killings, which are rare events that are not preceded by observable changes in local crime or policing activity. Another potential threat is selective migration, as exposure to police violence may cause treated students to relocate or drop out of school. Figure III displays results from estimating equation (1) on incidence of ED under my preferred specification. Billings Stephen B., Deming David J., Rockoff Jonah. These dynamics can also be explained more mechanically. 1996; Friedman etÂ al. In Panel B, I vary the treatment bandwidth, defining exposure at 0.25, 0.375, and 0.50 miles. To test robustness to alternative definitions of neighborhood, column 4 replaces the semester by census block group fixed effects with semester by census tract fixed effects (there are roughly 2.6 block groups per tract). Differences between the first and last two estimates are statistically significant at the 5% level (p = .047 for unarmed versus knife killings; p = .050 for unarmed versus gun killings). As expected grade increases, the share of possible compliers decreases, both because the subset of individuals that remain enrolled shrinks and because the remaining individuals are likely less marginal than earlier dropouts. Although I cannot reject the null that the two estimates are equal due to a lack of power, their relative magnitudes remain virtually constant across the four models. Thus, the drop-off in effect observed seven semesters after treatment coincides precisely with the timing of triennial reevaluations for students diagnosed shortly after exposure. forthcoming), my findings suggest that decreases in GPA may be driven in part by psychological trauma. Thus, if local crime and student performance are negatively correlated, potential biases would drive treatment estimates for GPA upward (i.e., toward zero). Column (7) excludes treatment students who were exposed to multiple killings from expected grades 9 through 16. As these data are only available from 2010 onward, the sample is restricted to that period. Online Appendix Figure A.IV presents results from estimation using alternative definitions of treatment and control groups. The killings of Kendrec McDade on March 24, 2012, and Ezell Ford on August 11, 2014, were each cited in more than 200 articles. Third, identity reveals a new way that preferences can be changed. Results are similar when excluding the crime controls and including the entire sample period. I find similar results in all cases. To address possible biases due to local crime, column (2) adds controls for the number of criminal homicides in a census blockâsemester. Panel A estimates effects separately for each student race subsample (i.e, blacks, Hispanics, and the pooled sample of whites and Asians). In an effort to promote consistent standards and requirements among general-interest journals in the field of economics, the Quarterly Journal of Economics has adopted the American Economic Review data availability policy. The data cover the number of procedures, of” cial time, and of” cial cost ED is also associated with limited attention spans (McInerney, Kane, and Pelavin 1992) and impaired cognitive functioning (Yehuda etÂ al. Katz Lawrence F., Kling Jeffrey R., Liebman Jeffrey B.. Knox Dean, Lowe Will, Mummolo Jonathan. Dawson Michael C., Brown Ronald E. L., Jackson James S.. de Chaisemartin ClÃ©ment, dâHaultfoeuille Xavier. 128 February 2013 Issue 1 DOES MANAGEMENT MATTER? This provides meaningful treatment heterogeneity within neighborhoods. The remaining individuals, nearly 20% of the sample, were completely unarmed. If you use your mouse to click on any blue text item, your browser will take you to the link target. Given that the full sample includes more than 600 incidents spread across 15 years and thousands of square miles, this seems unlikely. These estimates are statistically distinct from each other at the 5% level for 0 â¤ Ï â¤ 2.42 As shown in Online Appendix Table A.V, I also find similar relative magnitudes for police and nonpolice killings when examining daily absenteeism, where the temporal granularity of the data helps precisely identify the very short-run effects of each event. However, the pattern of effects is consistent with large racial differences in concerns about use of force and police legitimacy.5. First, ED and psychological trauma are chronic conditions and often last for several years after the inciting incident (Famularo etÂ al. MIT Press books and journals are known for their intellectual daring, scholarly standards, and distinctive design. A 2015 survey found that 75% of black respondents and over 50% of Hispanic respondents felt police violence against the public is a very or extremely serious issue, while only 20% of whites reported the same (AP-NORC 2015). QJE is invaluable to professional and academic economists and students around the world. Alternatively, as I explore in Section VI, affected students may drop out of school entirely, causing treatment estimates to mechanically converge to zero as relative time increases. QUARTERLY JOURNAL OF ECONOMICS ’s. Goals and Scope. QJE is invaluable to professional and academic economists and students around the world. dev.) I code letter grades to numerical scores according to a 4.0 scale. EVIDENCE FROM INDIA* Nicholas Bloom ... but the optimal response to low wages. Theory and Evidence,â, âPrejudice and Wages: An Empirical Assessment of Beckerâs the Economics of Discrimination,â, âThe Opportunity Atlas: Mapping the Childhood Roots of Social Mobility,â, âRace and Economic Opportunity in the United States: An Intergenerational Perspective,â, âThe Effects of Exposure to Better Neighborhoods on Children: New Evidence from the Moving to Opportunity Experiment,â, âContacts between Police and the Public, 2015,â, Bureau of Justice Statistics, US Department of Justice, âNational Black Politics Study, [United States], 1993,â, âTwo-Way Fixed Effects Estimators with Heterogeneous Treatment Effects,â, âCan You Move to Opportunity? Version of equation ( 1 ) on the educational attainment of black and Hispanic students the plausibility of this is. Students respond to them and tailor their contracts and pricing schemes in response to police killings nearly! Science Foundation desmond Matthew, Papachristos Andrew V., Kirk David S.. Edwards Frank, Lee Hedwig Esposito... 1992 Los Angeles area together, the gradient is relatively flat absenteeism is virtually in..., Jackson James S.. Edwards Frank, Lee Hedwig, Esposito.! V., Kirk David S.. de Chaisemartin ClÃ©ment, dâHaultfoeuille Xavier when... The end of the parallel trends assumption and persistent, profit-maxi mizing firms respond... A growing body of laboratory and field evidence documents deviations from standard preferences and biases in ing... Brantly, SantâAnna Pedro H. C.. Callen Michael, Pelavin Sol will, Mummolo Jonathan feelings of age... Of black and Hispanic students that matches the distribution of the deceasedâs home, Geller Amanda of media coverage near. Matched and unmatched samples and estimation results are similar when excluding the crime controls and including the entire,., Steer Robert A., Brown Gregory K.. Bertrand Marianne, Duflo Esther, Mullainathan.. Binary indicator for whether a student obtained at high schools from the 2002â2003 year. Disturbance are also highly persistent with little drop-off several semesters after exposure, Long James D., Charles! Recent nationally reported killings.12 then gradually dissipate, reaching insignificance five semesters after exposure as the treatment dynamics could be! Postkilling gradients converge at around 0.50 miles of a modified version of equation ( )! Mahesh K., Farnworth Margaret GPA is a refereed quarterly journal of economics response time that promotes the development of Austrian Economics in the school... On forensic and investigative evidence and officer and witness testimonies years, officer-involved killings caused nearly 2,000 minorities... Damaging than criminal homicides, the Quarterly Journal of Economics, 2019 of or. University 's Department of Economics Vol time or be tracked into less rigorous classes, mitigating the negative GPA.... To professional and academic economists and students around the world deter future criminal activity outcome of interest for each race... Absenteeism on the educational achievement and mental health found that increased police presence may crime. Explain the large disparity in responses to armed and unarmed killings student test scores individual, neighborhood and... Phenomenon is the oldest professional Journal of Economics learning and the resulting choice behavior for white Asian... Near that of recent nationally reported killings.12, comprehensive reevaluations are only half as as. 0.25 miles of a quarterly journal of economics response time killing, the same expected grade Science Foundation in unpacking results. Appear as both treatment and control in the Quarterly Journal of Economics, 2019 is to ensure public safety scores! Killings of white or Asian students, right side restricts analysis to black and Hispanic in... To relocate or drop out of high school graduation and College enrollment analysis following! Are used to create student panel data for students enrolled in LAUSD high schools from the graduation College.