This preview shows pages 1–8. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Discrimination II Lecture 7, February 28, 2012 Announcements Term Paper: First draft due Friday March 2, 10:00am. Be sure to test drive your peerScholar accounts well in advance. Userid : official ROSI email address; Password : utorid Remember: late submissions will not be accepted. Term Paper  Other deadlines: Peer Assessments, Tuesday March 13 th , 11:00 am Final Submission, Tuesday March 20 th , 11:00 am On peerScholar and turnitin. Final Exam: ALI TUE 17 APR AM 911 SEEL LLZ TUE 17 APR AM 911 SHER And dont forget to work on the problems. Tutorials resume this Friday, March 2 nd Alfia will be taking up selected problem set questions 1 Outline Review: Theories of discrimination Measuring discrimination Observability problem; Using regression analysis to detect discrimination in outcomes; The OaxacaBlinder Decomposition Derivation Numerical Example Empirical Application Review of evidence for men vs. women Discrimination by race and ethnicity Overview of policies towards discrimination 2 Review Theories of discrimination  how can discrimination be incorporated into economic models? Beckers model of employer discrimination: Based on employer prejudice (tastes) Based on customer or employee prejudice Implications of Beckers model Coexistence of discrimination and competition? Evidence that competition reduces evidence of potential discrimination Statistical Discrimination Employers infer individual productivity on the basis of group characteristics. 3 Review: Statistical Discrimination Individuals display a noisy signal of their true productivity: Such that: Employers will offer a wage based on expected productivity of the individual, given available information on the individuals noisy signal, and his/her membership in group x (A or B): This is a straightforward signal extraction problem. 4 i = i + u i u i N 0, u 2 ( ) E i i , x Case 1: Different mean productivity by group It is not difficult to show that the employers best prediction is given by: Where: Expected productivity is a weighted average of the average group productivity, and the individuals own noisy signal. The weight depends on quality of the signal: As , then , and is fully informative; As , then , and all conclusions are based on group means. 5 E i i , x = 1 ( ) x + i = x + i x ( ) = 2 2 + u 2 u 2 = = 1 i = u 2 Case 2: Different Signal Quality Assume now that the groups have identical average productivity: But that the signals are of different quality, i.e., We can show in this case that: For the group with the higher variance (poorer signal quality), the is less informative, and expected productivity is closer to the average. less informative, and expected productivity is closer to the average....
View
Full
Document
This note was uploaded on 03/30/2012 for the course ECO 339 taught by Professor Mbaker during the Spring '11 term at University of Toronto Toronto.
 Spring '11
 Mbaker

Click to edit the document details