Econ 1015 Final

total factor productivity

Y = f(A, K, L)

isolation index

a measure of exposure, minority weighted average of each section's minority population Sum [ B/Btotal * B/population]

patronizing equilibrium - danger of affirmative action

affirmative action can lead to this issue Employers have statistically correct beliefs about the inferiority of Bs and lowertheir standards for Bs so that they are assigned to the skilled task at the same rate for W workers This in turn further removes incentives for B workers to invest, completing the self-fulfilling prophecy. alternative policy: subside the assignment of black workers to skilled jobs, either by paying employers to do so or helping black workers assigned to the job

The Long Run Consequences of Living in a Poor Neighborhood, Oreopoulos (2003)

assignment to public housing experiment. treatment is being assigned to a project with high density of low income households (Toronto) threat to identification: families could reject first offer of housing without being taken off the list, but that probably didn't happen much individuals not in projects have significantly worse outcomes in worse neighborhoods. individuals in projects have similar income levels in better or worse neighborhoods

Causes and Consequences of Black Names (Fryer and Levitt 2004)

black names have positive effect for blacks and negative effect if given to white create a Black Name Index variables associated with low socioeconomic status are associated with more black names for all birth samples without controls, blacker names are correlated with worse life outcomes (makes sense bc correlated with lower SES). carrying a black name is more of a consequence than a cause

Salt Slavery hypothesis

blacks suffer disproportionately from cardiovascular deaths salt retention is big factor to cardiovascular issues. blacks may be more likely to have salt sensitivity gene because it would have helped them when they were sick and denied water on slave ships/plantations (perspiration, seasickness, dysentary) evidence supports the theory. Africans living in Africa have same levels of salt sensitivity as whites in the US, but blacks descended from victims of the slave trade have higher blood pressure than either group

odds ratio coefficients

coefficients from logistic regressions. gives percentage likelihood change in odds that outcome is 0 or 1

difference in difference estimate

compare outcomes in one set to outcomes in another set, to determine difference in outcomes relative to difference in sets

random forest

creates a large number of decision trees and averages over them. good for data if it has important interactions (collinearity) and non-linearities

Going Term Effects of Africa's Slave Trades, Nunn (2008)

cultural of mistrust, history of selling friends and enemies into slavery uses instrumental variables, distinguish between internalized norms and culture as cause and external environment and institutions internal norms: im[act of slave trade on individuals ancestors external environment: intensity of slave trade in geographic locations where individual is living today finds slave trade is negatively correlated with all five measures of trust Dunn uses distance from coast as an instrumental variable. Doesn't change the role of slave trade on influencing levels of trust. Also suggested slave trade affected trust through both external factors and internal norms. However, internal factors matter about twice as much.

From Separate and Unequal to Integrated and Equal? School Desegregation and School Finance in Louisiana (Reber 2010)

desegregation induced white flight to other neighborhoods and private school, but didn't change money going to schools (because property taxes barely changed or didn't) desegregation therefore did achieve its goal of improving the resources available in schools that blacks attended

Civil Rights, the War on Poverty, and Black-White Convergence in Infant Mortality in the Rural South and Mississippi (Almond, Chay, and Greenstone 2007).

desegregation of hospitals lead to far lower black infant mortality rates use many increasingly demanding tests to show causal relationship, ending with event study relationship that looks very convincing mandated integration accounted for 15% of the national decline in the black post-neonatal mortality rate

Desegregation and Black Dropout Rates (Guryan 2004)

differences in differences, look at groups outs in Birmingham vs dropouts in St Louis before and after Civil Rights act civil rights cases were focusing on case they could win, not those that had the biggest benefits. also study could be confounded by migration that happens as a result of cases (or result of lack of cases) uses IVs: 1) decade of first legal case filed for desegregation and 2) decade of first integration plan. showed delay not driving earlier estimates. therefore, desegregation lowered dropout rates in short term.

final conclusion about indices that determine income mobility (note: lots of overlap with indices that determine mobility out of poverty)

educational attainment resilience high quality parenting lack of problems with drugs.alcohol

measures of segregation

evenness: the differential distribution of two groups across areas of a city exposure: the amount of potential contact and interaction between members of different groups concentration: the relative amount of physical space occupied by a minority group centralization: event to which a group is located near the center of an urban area clustering: the degree to which geographic units inhabited by minority members about one another or cluster spatially

Ralph Bunche

extraordinarily successful academic, diplomat, professor still suffered from discrimination (barred from a tennis club baed on race)

The Effects of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports (Lochner and Moretti 2004)

find schooling significantly reduces the probability of incarceration. use Census and FBI data. reject tests for reverse causality

collinearity

high correlations among independent variables. problems of bias can make it hard to figure out what best variable to intervene on is solutions: 1. If we assume that the underlying structural model is linear in covariates,solution = Principal Component Analysis + Lasso (least absolute shrinkageand selection operator). 2. If we assume that the underlying model is non-linear in covariates, potentialsolution = Random Forests.

out of sample prediction

how well does the algorithm do on a sample with which its coefficients were not estimated

Stigma

idea from Loury "Race" as a social phenomenon results from two processes, "categorization"and "signification" Categorization is the process of sorting people into a finite number ofcategories (subgroups) based on bodily marks. Signification is the process of associating certain connotations with those categories. misattributed causality!

Omitted Variable Bias

if you don't include all the relevant variables, estimates will be biased. If you drop relevant variable X2 and only include X1, bias in coefficients occurs if Cov(X1, X2) != 0 and Beta2 != 0

Narrative of the Life of Frederick Douglas An American Slave,

importance of education harsh nature of slavery his mistress tried to reach him, but. was eventually forbidden to do so by her husband .then she sort of bought in. biggest impacts are on murder and assault rates. social savings could be 14-26% much of which comes from completing high school

Without Consent or Contract: The Rise and Fall of American Slavery (Fogel 1989)

intense economic analysis, look at how slavery was profitable. huge amalgamation of data. before Fogel and Engerman's book "Time on the Corss" was published it was widely believed that slavery was unprofitable and inefficienct. They showed it was actually hugely profitable for slave owners. duh. slaves worked extraordinarily hard and long hours, women worked right up until child birth, they cost way less than they produced (and pricing was relative to ability to produce) they made the south very wealthy and the practice would have continued if not for the civil war Fogel calculated total factor productivity for slave and non slave farms, finds that slave farms much more productive. Compare northern and southern farms. Huge jump in productivity for 16-50 slaves because of use of the Gang Labor system ( competition between teams of slaves, worked 75% more intensely per hour)

Guess Who's Been Coming to Dinner? Trends in Interracial Marriage over the 20th Century (Fryer 2007)

interracial marriage still uncommon. adjust for relative supply of members of various races in the US population. highly educated people far more likely to have interracial marriage use Oaxaca decomposition to decompose effects from different characteristics three theories 1. social exchange theories - individual characteristics each valued, have societal cost of marying a person with whatever set of characteristics 2. search/interaction models - just a matter of meeting people at random 3. marriage market model - marriage occurs when marrying increases total household output (includes like new social and environmental outputs) find that there is some sort of racial preference in dating, sue to same race preferences, nonrace attributes like education can't account for it

Instrumental variables

introduce variable Z that is relevant (correlated with X, Cov (X, Z) != 0 ) and exclusive (Z influences Y only through its relationship to X

Hatred and Profits: Under the Hood of the KKK, Fryer and Levitt 2012

it was a pyramid scheme, hugely profitable for those at the top. profits from memberships, robes, and taxes native born, low level professionals, white (duh) most likely to be members market shock, Stephenson trial, made it no longer cool to be in the KK, people stopped paying to be members

Teacher Incentives and Student Achievement: Evidence from New York CityPublic Schools (Fryer 2013)

let schools determine how incentives were distributed -> didn't help because schools tend to think about teachers not pupils. find no evidence that teacher incentives increase student performance, attendance, or graduation, nor do I find any evidence that the incentives change student or teacher behavior. If anything, teacher incentives may decrease student achievement, especially in larger schools.

dissimilarity index

measure of evenness (differential distribution) 1/2*Sum [ B/Ntotal - NB/NBtotal ]

mini coefficient

measure of inequality

a measure of segregation based on social interactions (Echenique and Fryer, 2007)

new method for describing segregation, social segregation indeed, that works at level of individual and thinks about interactions and who people people interact with interact with. new measure, SSI, satisfies 1. monotonicity 2. linearity 3. homogeneity

Cicala, Fryer, and Spenkuch (forthcoming)

new model of identity: self selection int opeer groups, perfectly competitive market for social groups. Equilibrium depends on supply and demand for nerds and troublemakers desire to achieve depends on what kind of rank student is trying to achieve

An American Dilemma (Myrdal 1947),

obstacles to African american participation in society worked with Bunche (Bunche didn't get enough credit) cited in Brown v Board groundwork for integration and affirmative action

Economics and Identity (Akerlof and Kranton 2000)

people fit into categories, which have inherent social norms and behavioral prescriptions. utility from informing armies identity. violating norms evokes discomfort and reduces utility, and also creates negative externality for others

confidence intervals

percent likelihood that the real Beta is within two standard deviations of estimate

theories of segregation (seen in Cutler, Glaser, and Vigdor)

port of entry theory: people prefer living with their own group centralized racism./collective action racism: whites keep blacks out of their neighborhood through legal/regulatory barriers and violence decentralized racism theory: whites prefer to live more with other whites and therefore pay more to segregate themselves. creates segregation premium

The Declining Significance of Race: Blacks and Changing American Institutions (Wilson 1978)

preindustrial period: slavery industrial period: race competition in economic life, white used black strike breakers, few protections for black workers, moderate prosperity increases but general discrimination. increase in racial solidarity and self sufficiency modern period: increased employment, some help from new deal (but proportionally less), improvements after WWII in occupation opportunities (higher paying openings), however still a black underclass, still a segmented labor market racial competition in economic matters/occupation is reduced in modern period relative to previous periods, but discrimination still persists

Racial Disparities in Job Finding and Offered Wages (Fryer et al 2013)

provides test for lower bound of discrimination in labor market estimates that discrimination accounts for 1/3 of black white wage gap atterns in our data are most naturally rationalized through a search-matching model in which employers statistically discriminate on the basis of race when hiring unemployed workers but learn about their marginal product over time

Intent to Treat Estimates

rather than randomizing D, offer the opportunity to participate in sub set and then look at compliance. Selection factors become self selecting

BM 2004

show evidence of differential returns to resume quality, no differences between range of resume characteristics but characteristics that are good for white resumes not always significant for black resumes test zip codes as well as names show black sounding names are not really more or less likely to get a call back from resume The only characteristics for which Black-sounding resumes are less likely toreceive a callback is whether the job is advertised by a federal contractor

he Strange Career of Jim Crow (Woodward 1955)

slavery and early reconstruction were marked by proximity between races in the south. north wasn't that friendly during this time either. Two "Southern" political views on Blacks during Redemption:Conservatives (Democrats): Blacks as inferiors. Whites as custodians thatmust lift Blacks up by granting them civil rights. Populists (Radicals, Republicans): Race was not dividing social factor, butsocioeconomic status. Blacks and poor whites both "in the ditch", andneeded to unite against wealthy conservatives. De jure segregation resulted from weakening of the forces restraining negrophobia: Federal Administration: subjugation of foreign countries provided ground fordomestic segregation. Compromise of 1877 and Plessy v. Ferguson(1896) taken as federal sanctionings of state-level, court-driven segregation. Southern Politics: conservatives and populists vied for control of the Blackvote share. When one party controlled it, the other recruited negrophobeelements to counteract it. Coupled with unfolding of corruption schemes andelectoral frauds, Blacks were framed as responsible for political impropriety inthe South. Parties united around negrophobe elements and need todisenfranchise Blacks. 1930s increase in activism. 1955 brown v. board of education, ended precedents from poesy vs Ferguson 1960s civil rights movement, MLK voting rights act 1965

Tournament Incentives in Chicago drug gang, Levitt and Venkatesh (2000)

tournament incentives to gang membership: large reward for moving up in the hierarchy, those who join gangs must not discount the future too heavily to be motivated by the incentives

Testing treatments

try to show causality by running an experiment where you compare treatment and control Regression recovers the treatment effect only when the selection effect is zero (when D is conditionally uncorrelated with potential outcomes, when D as "as good as randomly assigned" How to know D is as good as randomly assigned: 1. run an experiment (however this is not always possible or representative) 2. go find randomness in the real world

Are Ghettos Good or Bad? (Cutler and Glaeser 1997)

trying to figure out if segregation is good or bad find that housing segregation is highly correlated with black income segregation use difference in difference estimate, find that segregation is bad for blacks, has no effect on whites (or small positive) 2/3 cannot be explained by income, education, or commute segregation

Inputs and Impacts in Charter Schools: KIPP Lynn (Angrist et al 2010)

use lottery strategy, students grades 5-8, no excuses approach huge achievement increases in math, reading, and science

Forty Years of Economic Progress for Blacks (Smith and Welch 1986)

wage difference decreasing, opportunities for blacks increasing slowly

The Letter from Birmingham Jail

written by MILK (preacher, activist) he had been jailed for taking part in a March. wrote line from jail, "injustice anywhere is a threat to justice everywhere"

Engerman-Sokoloff Hypothesis (2002)

"extreme differences in inequality that arose in the early New World due to the extent of slave use may have contributed to systematic differences in the way institutions developed - which would affect growth today." economic inequality makes institutions cater to elites, so they don't provide good foundations for economic growth plantation systems, especially gang labor, led to institutions based in inequality (CHECK WHAT PAPER) found negative relationship between plantations in the past and current income. Not just in US, but also in West Indies, other places that had plantation slavery systems Other studies found that all types of slavery in the past have negative impact on current income. Actually, some found evidence against the Engerman-Sokoloff Hypothesis, found places with non plantation slaves and even more significant negative impact

Letter from a Birmingham Jail (King 1963)

"injustice anywhere is a threat to justice everywhere"

facts about US crime

-most criminals are young men, especially less educated minority men -US has highest incarceration rate in the world -higher murder rates than most developed ocuntries -increase in crime rate in 1960s -decrease in crime rate since 1990s -incarceration has increased a lot since 1980 -crimes mostly concentrated in cities

Evidence of Racial Bias1

1. Are Emily and Brendan More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination (Bertrand and Mullainathan 2004) 2. Discrimination in a Low-Wage Labor Market: A Field Experiment (Pager et al. 2009) 3.What Happens Before? A Field Experiment Exploring How Pay and Representation Differently Shape Bias on the Pathway into Organizations (Milkman et al 2015) 4.Racial Discrimination in the Sharing Economy: Evidence from a Field Experiment (Edelman, et al 2017)

Principal component analysis with lasso

1. aggregates correlated variables 2. finds a hyperplane that explains most of the variation in the data 3. finds the next aggregate that is linearly orthogonal to the previous aggregate that explains most of the variation assumes linear model inn aggregates

channels for difference in 6 year black white life expectancy difference

1. resource differences 2. health behaviors 3. environment - physical and psychological 4. genetics and the legacy of slavery (truncating population distribution)

Long data vs Wide data

Bid Data is long (Chetty 2018 is an example) wide data not necessary to study mobility. width more important. long and wide is best. long and narrow is bad - could be many confounding covariates left out

The Effect of School Choice on Participants: Evidence from Randomized Lotteries (Cullen, Jacob, and Levitt 2006)

Chicago school lottery program. students who win lotteries attend better schools. however, no evidence that winning the lottery provides any systematic benefit \. lottery winners have improvements on non traditional outcome measures (lower arrest rates)

Becker distinction between discrimination and segregation

Discrimination arises when B and W labor (capital) receives different wages(returns on capital) despite being equally productive. Both W and B laborers(capitalists) participate in the same market. Segregation arises when B and W do not participate in the same markets.This also includes the case in which tastes for discrimination impede B fromparticipating in the same markets.

Levitt (2004) explanations that don't work for 1990s crime decline and 4 that do

Dont: 1. strong economy of 1990s 2. changing demographics, fewer adolescents but decline smaller than that in 1980s 3. better politicking strategies 4. gun control laws 5. conceal and carry laws 6. increased use of capital punishment Do: 1. increases in the number of police 2. increased incarceration (though somewhat of a timing problem) 3. decline of crack epidemic, crack no longer as lucrative for gangs (Fryer 2013) 4. legalization of abortion also, potentially childhood lead poisoning and crime (Reyes, 2007)

Devil in a Blue Dress (Mosley 1990)

Easy Rawlins goes from day laborer to detective. woman is passing for white to protect her lover's reputation

Parental Incentives and Early Childhood Achievement: Evidence from Chicago Heights (Fryer, Levitt, and List (2015)

Intent to Treat example. Let parents decide to either get cashier child's academic improvement or have it put in a trust for college large and statistically significant positive impacts on both cognitive and non-cognitive test score of hispanics and white, but no impacts on blacks (very heterogenous effects!) difference not attributable to observable characteristics or to intensity of engagement.

The Medium-Term Impact of High Achieving Charter Schools (Dobbie and Fryer 2013)

Look at HCZ, see if test score gains translate to improved life outcomes 14.1 percentage points more likely to enroll in college. Admitted females are 12.1 percentage points less likely to be pregnant in their teens, and males are 4.3 percentage points less likely to be incarcerated. no changes in self reported health but students weren't that old when the survey was administered high-performing schools may be sufficient to significantly improve human capital and reduce certain risky behaviors among the poor.

Police Brutality and Crime: Coming of Age During the Rodney King Riots (Fryer 2018)

Look at impact of first viral shooting (Rodney King) find 6 year spike in total crime, more officers killed or assaulted, no real change in wages or education attainment check

hypothesis testing

Null hypothesis, H0, is baseline value for Beta if it would be surprising that, given our results, the null hypothesis is true, we reject the null hyptoeshsis p-value is probability that, under H0, we would see a test statistic at least as extreme as the one observed. Smaller p value -> greater confidence that the null hypothesis is wrong

Road map to data driven interventions

Phase I: Conduct in-depth surveys and face-to-face interviews with arepresentative group of adults who were raised in poverty. In the surveys, weask adults to reflect on their childhood and family behaviors, attitudes and experiences. Phase II: Use the data to identify which correlates, and in what relativeimportance, differentiate individuals who escape poverty from those who stay trapped. Phase III: Implement a data driven intervention to test the effectiveness ofinterventions designed to impact these correlates.

Nothing Impossible (Monroe 1999)

The Monroe Doctrine: leadership method used by Monroe to turn aroundfailing schools Frederick Douglass Academy: turned the school around, uniforms, captive lunch, better teachers, tutoring, made it best school more time in Schoo, tutoring, manage teachers, data driven instruction, high achieving culture

Does Competition Among Public Schools Benefit Students and Taxpayers? (Hoxby 2000)

Tiebout Choice = a term for a class of economic models in which agents'choices of where to live determine the supply of schools Tests whether increased competition between schools leads to increasedstudent achievement and/or decreased per-pupil expenditures measures the degree of tie bout choice. finds that increased competition relates to more productive schools, however later studies find a lot of problems. Rothstein (2007) tries to replicate Hoxby's results, but has a lot of trouble. unclear if Tiebout choice actually leads to increased achievement. However, per pupil expenditure results hold up well

Market Wages and Youth Crime (Grogger 1998)

Trade off between crime and legitimate labor implies that as legitimate opportunities change, crime should to. Finds that 25% of racial difference in criminal participation rates can be explained by the black-white wage gap also explains why age associated with decreased crime, because increased wages means increased pay in legitimate labor force

Understanding the Black-White Test Score Gap in the First Two Years of School (Fryer and Levitt 2004)

racial gap in cognitive testing emerges as early as 2 years old most important control is number of children's books in children's home most likely reason for difference is that on average blacks attend worse schools than whites. there is no racial difference between students who attend integrated schools or who are in the same schools. blacks also suffer worse summer setbacks

Crime control policies

1. incapacitation - keep criminals off the street 2. deterrence - make participating in crime less appealing via sanctions 3. prevention - make sure that legitimate work pays better 4. rehabilitation - improve opportunities for ex-offenders

factors for analyzing literature

1. reliability 2. representativeness 3. endogeneity 4. counterfactuals 5. covariates

Harlem Childrens Zone

97-block area in Harlem that combines "No Excuses" charterschools with a web of community services designed to ensure the socialenvironment outside of school is positive and supportive for children frombirth to college graduation

Management and Student Achievement: Evidence from a Randomized FieldExperiment (Fryer 2017)

Across two years, principals were provided 300 hours of training on lesson planning, data-driven instruction, and teacher observation and coaching. The findings show that offering management training to principals significantly increases student achievement in all subjects in year one and has an insignificant effect in year two. results in year two might be a result of principle turnover, as well as cumulative nature of training school with principles that remain have treatment effects in both years (especially if principles are predicted to continue to implement training) when principles are predicted to leave, effects are statistically insignificant in each year of treatment principle enforcement is super necessary! Potential mechanisms: 1. placebo effect 2. learning curve for new principals 3. complementarities in content and systems management

Implicit social cognition

Banajio and Greenwald (1995) ttheory to explain many instances of "unwilling" or "unconscious"associations people make, including those with implications for racial discrimination. the cognitive process through which pastexperiences influence actions in social interactions in a way that is notperceived by the actor. In other words, the agent does not know or realizethat past experiences are shaping his current behavior.

Racial Bias in Motor Vehicle Searches (Knowles, Persico, and Todd 2001)

Black motorists more likely to have their cars searched in the US. There have been several lawsuits re racial profiling. Beckerian outcomes test for discrimination (Becker made a model to look for discrimination in mortgages) find that police made choices based on success rate of subgroups. No racial prejudice in Maryland, and if police have utility only for large drug finds, theres a bias against white drivers.

Economics of Discrimination (Becker 1957)

Created Discrimination Coefficient, tax on economic agent that discriminated (or were discriminated against) captured discrimination as part of net transaction costs in labor market (which previously people thought was just about money costs) Market Discrimination Coefficient measures the proportional difference in white Black wages in the real market model of two societies, blacks and whites. without discrimination each laborer paid marginal product, but if one has taste for discriminating it changes wages. Exported W capital must receive a larger return than domestic W capital,because it has to be willing to be combined with B labor.But then, W capitalists could stop providing capital to the W society andprovide capital to the B society, in order to earn more.Thus, the net return on all W capital (after accounting for discriminationtastes) has to be the same as the return on domestic W capital.Since the return on exported W capital is smaller than the net return onexported W capital, exported W capital to B decreases.But then, there is less W capital to combine with B labor, which in turnreduces B labor income. Therefore, discrimination hurts whites and blacks, and blacks don't have a an incentive to retaliate difference between discrimination (same market) and segregation

Getting Beneath the Veil of Effective Schools: Evidence from New York City (Dobbie and Fryer 2013)

Find that traditional input measures to charter schools (class size, per pupil expenditure, fraction of workers with no certification, and fraction of teachers with an advanced degree) are not correlated with school effectiveness find that five policies do help though 1. human capital management, frequency of teacher feedback 2. data guided instruction 3. high dosage tutoring 4. increased instructional time 5. high expectations explains 45% of variation!

Fist Stick Knife Gun: A Personal History of Violence (Canada 2010)

Gang violence was increasing and becoming more violent due to drug trade and availability of guns. Schools need to be stricter on guns, as does the government. Social and parental pressure to respond to violence by fighting back. Canada is former CEO and current president of HCZ

The Fire Next Time (Baldwin)

It contains two essays: "My Dungeon Shook — Letter to my Nephew on the One Hundredth Anniversary of Emancipation," and "Down At The Cross — Letter from a Region of My Mind." The first essay, written in the form of a letter to Baldwin's 14-year-old nephew, discusses the central role of race in American history. The second essay deals with the relations between race and religion, focusing in particular on Baldwin's experiences with the Christian church as a youth, as well as the Islamic ideas of others in Harlem.

The Impact of Federal Investigations on Crime and Policing (Devi and Fryer 2018)

Look at Pattern-or-practice investigations by federal gov into police department, how they impact practices. started 1994 used to root out discrimination in p[olicing use differences in differences and cumulative difference to figure out results of investigations on crime rates during the investigation look at 2013 Martin killing, BLM movement, event study. role of viral shootings look at changes in community engagement and trust (911 calla) , changes in policing (911 response time and civilian contact) cannot reject the hypothesis that sudden media attention decreasescontacts for all post-investigation months Consent decrees post-#BLM have become more aggressive in terms of thetotal number of pages devoted to reform practices, as well as the number ofsections dedicated to constitutional traffic and pedestrian stops post BLM, investigations increase homicide rates, rather than decrease - lives lost. probably due to shift in policing though its unclear policing might have changed due to media attention

The Pupil Factory: Specialization and the Production of Human Capital (Fryer, 2016)

Look at teacher specialization in Texas elementary schools found small negative impact of specialization in elementary schools, especially for students in special education consistent with a model where benefits of specialization are driven by comparative advantage of subjects - fewer interactions leads to inefficient pedagogy

Chetty et al 2011 and Chetty, Friedman, and Rockoff (2011)

Look at theimpact of class size, teacher experience, class quality and teacher value-added on wages. find increased in teacher quality (measured by experience) leads to increase of $1000 per student per yield, 16,000 NPV gain overall, equivalent to 6% raise at age 28

An Empirical Analysis of Racial Differences in Use of Force (Fryer 2016)

Looks at NYPD stop and frisk data, police public contact survey, , district level data, office shooting data. lots of issues with each set. results focus on extensive margin: whether or not an officer used a particular type of force on a suspect so far have found there are differences in non lethal use of force. as force increases this persists, but doesn't persist in officer involved shootings race coefficients seems to be larger among uses of force that are least penalized/observed by superiors

Racial Inequality in the 21st Century: The Declining Significance of Discrimination (Fryer 2010)

Racial differences in social and economic outcomes are greatly reduced when one accounts for educational achievement; therefore, the new challenge is to understand the obstacles undermining the development of skill in black and Hispanic children in primary and secondary school. the racial achievement gap is remarkably robust across time, samples, and particular assessments used. The gap does not exist in the first year of life, but black students fall behind quickly thereafter and observables cannot explain differences between racial groups after kindergarten several programs -- various early childhood interventions, more flexibility and stricter accountability for schools, data-driven instruction, smaller class sizes, certain student incentives, and bonuses for effective teachers to teach in high-need schools, which have a positive return on investment, but they cannot close the achievement gap in isolation. More promising are results from a handful of high-performing charter schools, which combine many of the investments above in a comprehensive framework

Estimating the Returns to Urban Boarding Schools: Evidence from SEED (Curto and Fryer 2014)

SEED schools - no excuses charter and five days a week boarding program show that SEED schools lead to increases in easing and math (large impacts!) but the returns on seed schools are low because of the costs of boarding. similar charter schools without boarding have greater returns on investment. maybe community isn't that necessary gains much larger for females - maybe because they're escaping abusive circumstances. they drive the gains in reading.

The Economics of Crime (Becker 1968)

Shows that crime can be modeled as rational, utility maximizing behavior. Shows that individuals commit crime if the pay off (discounted by risk) is greater than what they would earn in the traditional labor market Looks at expected financial gains, expected legal sanctions, psychic costs/stigma, and expected income from legitimate job (opportunity cost) Freeman (1999) wrote up a criminal participation decision equation Cewrtainty vs Severity of Punishment According to mode, raising legal sanctions a lot for even moderate crimes would end crime. However, raising p is costly, expensive and going to be unpopular (ie death penalty for littering)

Morris, S., 2001. Political correctness. Journal of politicalEconomy,

The dangers of political correctness Two agents - informed social scientist and uninformed policy maker The social scientist is advising the policy maker on whether armative actionmitigates racism and she has some noisy information on what the true stateof the world is. The policy maker and social scientist meet repeatedly to discuss issues onrace and solutions to them. Observation problem: the policy maker doesn't want to appear racist, therefore might choose a sub optimal policy Say, the social scientist observes that armative action does not mitigateracism. She can recommend anti-armative action policies but that makesthe policy-maker believe that the scientist has a higher chance of beingracist. If the scientist wants to be a part of future policies on race she liesand proposes armative action.As the policy maker also knows that non-racist social scientists might lieabout the true state of the world, they disregard all information from allsocial scientists.Political correctness leads to the loss of socially valuable information.

The Philadelphia Negro: A Social Study (Dubois 1899)

The first sociological study of the black community in the United States One of the first works in sociology to complement observation with data Stressed the importance of class and structural inequities in creating the Negro Problem, in a time when others were using calipers and rulers toclassify races (with blacks at the bottom of the biological hierarchy) to justify discimination Also emphasized the necessity for members of the black community torecognize and take ownership of their own problems description of the seventh ward. The Ward was incredibly diverse: on its western fringe lived affluent whites,at its center lived one of the nation's densest concentrations of black elites,and a diverse mix of the poor from both races lived on the Ward's easternfringes, which was also the site for the city's most notorious black ghetto. population of freedmen was very young. notes insufficient high paying employment opportunities for men. women needed to work, often had to work in distasteful fields. He recognizes the church as a source of community and, perhaps moreimportantly, as insurance in times of need Du Bois relates the high rate of crime in the black community to the rapidurbanization of freed slaves after the War, lack of economic opportunity,illiteracy, and the large portion of young, single men in the population. 42% repeat offenders.

Intergenerational Transfers and the Distribution of Wealth (Loury 1981)

allocation of investment in offspring cannot be a result of parents borrowing, thus its representative of their wealth redistribution doesn't conflict with economic efficiency check

Geography and Racial Health Disparities (Chandra and Skinner 2004)

blacks receive worse healthcare, tough maybe no longer due to discrimination different mechanism for disparities; African-Americans tend to live in areas or seek care in regions where quality levels for all patients, black and white, are lower. Thus ensuring equal access to health care at the local or hospital level may not by itself erase overall health care disparities. However, reducing geographic disparities in both the quality of care, and the quality of health care decisions by patients, could have a first-order impact on improving racial disparities in health care and health outcomes.

Housing Segregation, Negro Employment, and Metropolitan Decentralization (Kain 1968)

channels through which housing segregation may effect black employment: -commuting costs -segregation prevents information access -employers in predominantly white neighborhoods might feel social pressure to discriminate against black workers also potential for reverse causality - differential job opportunities could cause segregation find a large and significant effect of distance from an area on percentage of black employment in the area. lots of potential for selection issues though.

Accountability and Flexibility in Public Schools: Evidence from Boston Charter and Pilot Schools (Abdulkadiroglu, et, al. 2011)

charter school effect on achievement in Boston also look at Pilot schools (independent, but allow for some bargaining provisions for teachers) Lottery estimates show large and significant score gains for charter students in middle and high school. In contrast, lottery estimates for pilot school students are mostly small and insignificant, with some significant negative effects. Charter schools with binding assignment lotteries appear to generate larger gains than other charters. J

coefficients in regressions

coefficients measure the increase in Y associated with a 1 unit increase in the corresponding X

Divergent Paths: Structural Change, Economic Rank, and the Evolution of Black-White Wage Differentials, 1940-2014. (Bayer, P. and K. Charles. forthcoming)

confirms results from Smith and Welch, and then look at more recent data. also look at working and non working wages, and wages vs earnings black incarceration and out of labor force rates have increased dramatically relative to whites. level and rank gaps: blacks have higher value in black ranking because less competition (people with college degrees) rank gaps arise because blacks don't look as qualified rank wise when placed in white skill distribution jnrelative to black skill distribution. level gap: estimate the individual level quantile regression for each time period rank gap: estimate quantil regression, tells position of individual in the white earning distribution earnings gap arises because black and whites with the same skill level are rewarded differently distributional convergence is changes in how labor market rewards skill, leaves black and white positions unchanged in respective earning distributions. Positional convergence, however, is race sensitive. Object of interest, captures changes in black skill/return to black skill

The Anatomy of Racial Inequality (Loury)

cycle of tainted social information that has resulted in a self-replicating pattern of racial stereotypes that rationalize and sustain discrimination. need race based intervention, but not affirmative action because that would feed into cycle

The Plight of Mixed Race Kids (Fryer et. al 2012)

discrimination models: 1. discrimination based models: lower discrimination leads to more investment in human capital 2. conformity models : individuals have preferences over behaviors but also care about popularity and social esteem 3. two-sector roy model: individuals have preferences over behaviors but also care about popularity Roy model: self selection, teens select into black or white peer group. mixed race kids engage in riskier behavior because both groups hold them to higher standards

Self-Censorship in Public Discourse: A Theory of "Political Correctness" and related phenomena (Loury 1994, Rationality and Society)

discusses how people reading public messages don't trust senders of messages, and therefore senders edit their messages to cater to popular opinion. this can mean otherwise well informed people send messages that are less accurate than they could be. creates over reliance on euphemism and platitudes. was written when he was still against affirmative action. eventually realized that it might kind of be necessary

Mary McLeod Bethune: Building a Better World (Thomas, et al 2002)

education as key to racial uplift, emphasis on vocationalism (sort of like BT Washington) importance of women learning how to create a good home life too Mary also lead a number of policy organizations for women of color. highest ranking black women in the government at the time

Gatreaux Program Duncan, 2006

established in the late 1970s as part of acourt-imposed public housing desegregation remedy allowed black families to move to suburban areas, relocate "moving to opportunity" see that overall there is a treatment effect. adults don't duo much better in labor market or in health. Girls improve in math and reading a bit, boys do slightly worse (more stressed and get into more trouble) good for children under the age of 13 - increases college attendee and earnings and decreases single parenthood rates might be worse for older children because they lose friends might be better for girls because they leave abusers

Are High Quality Schools Enough to Increase Achievement Among the Poor? Evidence from the Harlem Children's Zone (Dobbie and Fryer 2011)

evaluating Promise academy in HCZ use intent to treat, instrumental variables (winning the lottery as instrument for years spent in the charter school and estimate the effect of attending the charter school) compare children living in the zone vs those not living in the zone suggest that the effects of attending an HCZ middle school are enough to close the black-white achievement gap in mathematics. The effects in elementary school are large enough to close the racial achievement gap in both mathematics and ELA. We conclude with evidence that suggests high-quality schools are enough to significantly increase academic achievement among the poor. Community programs appear neither necessary nor sufficient.

The Role of Pre-Market Factors in the Black-White Wage Gap Neal and Johnson (1996)

find that controlling for aptitude scores, black and hispanic women earn more than white women (may be due to selection bias in labor market) hispanic men have lower returns than those of everyone else find no statistically significant difference between AFQT score return between blacks and whites find evidence for black white gaps in pre market skills, from AFQT scores after controlling for pre-market skills (notinnate ability), racial wage gaps are sharply reduced for males, and oftenreversed for females. affirmed by Fryer (2010)

Reconciling Results on Racial Differences in Police Shootings (Fryer 2018)

find that there are not really racial differences in police shootings given context however the narrative of racial bias in shootings is widespread

Long-run Impacts of School Desegregation & School Quality on Adult Attainments (Johnson 2015)

follow up on Guryan (2004) controls for lots of trends, like school spending, avg income of students, etc plots treatment effect for blacks and whites before and after desegregation of their schools finds that time does matter - some effects of desegregation were part of pre existing trends. however, desegregation did still lead t small positive effect on blacks adult outcome s(an no effect on whites) also looked at siblings in different schools - controls for alot of factors. finds that desegregated schools meant better education, economic, and health outcomes for black children. school spending effects for blacks but not for whites

Political Economy of Hatred (Glaeser 2005)

hatred is the result of an equilibrium where politicians supply stories of past atrocities in order to discredit the opposition and consumers listen to them. The supply of hatred is a function of the degree to which minorities gain or lose from particular party platforms, and as such, groups that are particularly poor or rich are likely to be hated limiting anti minority policies will limit hate demand for hatred falls if consumers interact with hated group (unless interactions are primarily abusive)

The Autobiography of Malcolm X (X and Haley)

journey about being black in America, understanding muslim religion, roots of black power movement.

Craig and Fryer (2018) two sided statistical discrimination setting

like Coate and Lpoury for workers, but focuses on investments in amenities to accommodate minorities Craig and Fryer first show that, in the absence of a policy, it canneverbe thecase that both B and W invest and employers believe B and W to be equallyqualified - that is, there is no equlibrium with homogeneous beliefs !The intuition is simple: since Bs are a numerical minority in the model, it isalways costlier for firms to invest in amenities for Bs. Bs then have incentivesto invest at a lower rate than Ws, all else fixed. In turn, this creates an incentive for firms to assign less B workers to theskilled task. Futilirt of affirmative action: can't move us to an equilibrium with homogenous beliefs. There is no one sided policy that eliminates discrimination and doesn't run into the risk of triggering less investment by black workers policies must be two sided!! Must change one of the following 1. employer beliefs 2. employee standards for accepting workers, and those workers beliefs about the fraction of firms investing in hiring them 3. employer beliefs about the fraction of workers qualified and standards for applying to employers

Charter Schools and Labor Market Outcomes (Dobbie and Fryer, 2017)

look at Charter schools in Texas. find that, at the mean, charter schools have no impact on test scores and a negative impact on earnings. No Excuses charter schools increase test scores and four-year college enrollment, but have a small and statistically insignificant impact on earnings, while other types of charter schools decrease test scores, four-year college enrollment, and earnings. find that charter schools that decrease test scores also tend to decrease earnings, while charter schools that increase test scores have no discernible impact on earnings. In contrast, high school graduation effects are predictive of earnings effects throughout the distribution of school quality. potential mechanisms 1. age of the sample and returns to early work experience might skew low, but later earnings might be higher 2. high dropout rates in high performing charter schools 3. multi tasking, charter schools might substitute away from teaching skills that are valuable in labor market to teaching skills that are valuable for scores

Testing for Racial Differences in the Mental Abilities of Young Children (Fryer and Levitt, 2013)

look at mental aptitude tests of very young children no racial test score gaps in first year of life, but gaps emerge quickly with age

Financial Incentives and Student Achievement: Evidence from Randomized Trial (Fryer 2011)

looked at intent to treat effects: difference in outcomes for students who were supposed to receive treatment vs those who weren't supposed to differences by subgroup are insignificant, except for language proxies input incentives more effective than output incentives (lack of knowledge of academic production function, handles self control problems, complementary inputs, unpredictability of outputs - students might not realize they can control outputs) aligning incentives generated large outcome increases

Injecting Charter School Best Practices into Traditional Public Schools: Evidence from Field Experiments (Fryer 2014)

looking at best practices from high performing charter schools and moving them into low performing schools in Texas 5 practices: teacher feedback, data driven instruction, tutoring, more hours, high achieving culture found significant increases in Houston math scores at all levels, little to no effect on reading though saw similar results in Denver and Chicago

School Segregation, Educational Attainment, and Crime: Evidence from the End of Busing in Charlotte-Mecklenberg (Billings, et al 2014)

looking at results of ending race based busing, mechanical re zoning, no priority to poor students in lotteries. increase in segregation after 2002 look at variation in exposure to schools that vary in racial composition. students treated with the change in racial composition in 2002. potential for attrition bias, pre existing trends. find other reactions, like families moving or switching to private school, don't matter shifts resource allocations away from poorer schools, also changed peer to peer interactions.

Causal effects of incarceration on Crime Rates, Levitt 1996

looks at effect on crime of releasing prisoners early due to overcrowding. finds violent crime would have been 70% higher in 1996 if the increase in prisoners from 1971 to 1993 hadn't occurred estimates that the social benefit from an additional prisoner is 53900 per year. This is greater than the cost of incarceration, 25000-35000(not complete though because it doesn't include wasted human capital and pain and suffering)

Soul on Ice (Cleaver)

man's journey becoming a dealer and then rapist and part of black power movement and marxist revolutionary. arrested for involvement with white women.

Tribute to David Blackwell

math professor he was the first African American inducted into the National Academy of Sciences, and the first black tenured faculty member at UC Berkeley check

The Effects of Police Violence on Inner City Students (Ang 2018)

police violence has negative effects on GPA (2-4%, accounts for 86% of difference between average and minimum for college admissions) of students with same race of victim students living near incident obtain .25 years less of education, 15% less likely to graduate, 18% less likely to enroll in college

The Impacts of Neighborhoods on Intergenerational Mobility I: Childhood Exposure Effects (Chetty and Hendren forthcoming, QJE)

see lecture 22! moving children before the age of 13 is good for them, after is only a little good or even bad. $3,000 income increase - is that enough to claim victory?

Is School Segregation Good or Bad? (Echenique, Fryer, Kaufman 2006)

segregation is bad. use individual measure of social connections, but that has less of an important relationship with achievement find that within school segregation outcomes aren't causal friendships may not be the only relevant cross race social interaction that occurs within a school

Arrow (1973)

statistical discrimination employers dealing with prior beliefs and signaling to try and get best candidate for job if for any reason, employers believe W workers are more qualified than B workers, than wages for W workers are higher than wages for B workers=⇒Discrimination! beliefs and actions need to be self fulfilling statistica discrimination is a self fulfilling prophecy

How Does Peer Pressure Effect Educational Investments (Burstyn and Jensen 2015)

students more likely to publicly sign up for SAT prep in honors class than while in non honors class. peer pressure is powerful! changing social norms is difficult, but changing observability could help (less public recognition for academic achievement)

Native Son (Wright)

tells the story of 20-year-old Bigger Thomas, an African American youth living in utter poverty in a poor area on Chicago's South Side in the 1930s. While not apologizing for Bigger's crimes, Wright portrays a systemic inevitability behind them. Bigger's lawyer, Boris Max, makes the case that there is no escape from this destiny for his client or any other black American since they are the necessary product of the society that formed them and told them since birth who exactly they were supposed to be. "No American Negro exists", James Baldwin once wrote, "who does not have his private Bigger Thomas living in his skull."

F test

tests for whether characteristics join explain the values of Y

Will Affirmative Action Eliminate Negative Stereotypes (Coate and Loury 1993)

use a job assignment model to explore howstatistical discrimination can arise, and whether anti-discrimination policiescan help mitigate statistical discrimination. We suppose that the employer has aprior beliefthat B or W workers arequalified. Prior beliefs are probabilities. incorporate worker signaling, gives employers opportunity to update beliefs workers have to decide whether or not to invest in signaling, sending same signal has different costs for blacks and whites show that there can be multiple equilibria, beliefs about whites and blacks can be self confirming and result in discrimination against blacks affirmative action can be modeled as a constraint that makes employers assign blacks and white workers to the qualified task at the same rate, creates a price of equality

Forty Years of Economic Progress for Blacks (Smith and Welch 1986).

things look good for blacks at top of income distribution but little progress for the rest. schooling quality for blacks has improved looked at 1940-1980 data, find that although wages are converging theres a considerable gap. Blacks have greater returns to schooling (except in the most recent years). could be explained by blacks migration north during Jim Crow era, North had higher wages generally ("great migration"). however could be offset by larger gap in south. living in the south hurts black wages more than white wages. other mechanisms,s: agriculture in the south, affirmative action, declining black labor force participation

The Economics of 'Acting White' (Fryer and Torelli 2010)

two audience signaling problem: behaviors that promote labor market success are same behaviors that can induce peer rejection for black kids "acting white" Social Status Index - result of how many people list that person as a friend weighted by status of friends find negative relationship between popularity and grades for hispanics, positive until 3.5 for blacks then negative, overall positive for whites. negatives more dramatic for boys in all races potential reverse causality problem conclude that there are racial differences in the relationship between peer group acceptance and academic achievement differences are exacerbated in areas with more interracial contact or increased mobility

Becoming Richard Pryor

uncompromising examinations of racism and topical contemporary issues, which employed vulgarities and profanity, as well as racial epithets. He reached a broad audience with his trenchant observations and storytelling style, and is widely regarded as one of the greatest and most influential stand-up comedians of all time.

Fryer, Pager, and Spenkuch (2013)

use data from theSurvey Research Center at Princeton University on unemployment insurancerecipients to estimate Black-white wage gaps in both offered wages and jobsearch behaviors. test for wage discrimination establish: 1. blacks incur larger wage losses than whites due to job separation 2. blacks have lower reservation wages 3. blacks have higher returns to tenure within firms

The Rise and Decline of the American Ghetto (Cutler and Glaeser 1999)

uses dissimilarity and isolation indices to trace urban segregation Pre civil war: high dissimilarity, low isolation. One ghetto After second great migration, increased segregation. Segregation peaks in 1960s After 1970: increases in black population increases isolation, but not dissimilarity

Pay, Reference Points, and Police Performance (Mas, 2006)

uses final offer arbitration of wage disputes to study therelationship between pay raises, expectations, and employee performance Mas shows that final offer arbitration can have inefficiencies arising from thebehavioral response of participants to unfavorable outcomes Per capita number of crimes cleared by arrest is 12 percent higher in themonths following arbitration when arbitrators ruled in favor of police officers,relative to when arbitrators ruled for the municipal employer

Using Electoral Cycles to Estimate the Effect of Police on Crime (Levitt 1997)

uses timing of mayor and governor elections as an IC to identify causal effect of police on crime. police force sizes increase during governor election years. increases in policy reduce violent crime but have a smaller impact on property crime. we see that its possible that the marginal social benefit of reduced crime is equal to the costs of hiring additional police

conditional expectation function

what we would expect about Y given what we know about X's can be modeled as a linear regression