LN+11+Diff+in+Diff

LN+11+Diff+in+Diff - Empirical Methods II (API-202) Kennedy...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
Empirical Methods II (API-202) Kennedy School of Government Harvard University 1 Lecture Notes 11 Difference in Differences So far we have seen three ways of dealing with OVB : o Include omitted explanatory variables o Randomized experiments o Fixed Effects (FE) In the next classes we will see two additional techniques: o Today : Difference in Differences (DID) o Next week : Instrumental Variables (IV) I – DIFFERENCES IN DIFFERENCES We will illustrate the use of this technique with a study that generated a lot of controversy among economists. “Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania”, David Card and Alan B. Krueger, The American Economic Review , Vol. 84, No. 4 (Sep., 1994), pp. 772-793. o Policy question: What is the effect of the minimum wage on employment? o Standard economic theory predicts that this relationship is negative (i.e. as the minimum wage increases, we expect employment to decrease) o Card and Krueger tried to assess whether this prediction of economic theory was true empirically (i.e. using data from the real world) and to determine the magnitude of the effect
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Empirical Methods II (API-202) Kennedy School of Government Harvard University 2 Background o Survey of fast food restaurants in NJ and PA. (Wendy’s, Burger King, Roy Rogers) o NJ raised its minimum wage from $4.25 to $5.05 per hour (1992). PA stayed fixed at $4.25. o The paper uses PA as “control” and NJ as “treatment”. The argument is that they are very similar states (also share a border) so PA looks like NJ would have looked like without the change in minimum wage. In other words, PA is mimicking the counterfactual for NJ. Here is a summary of the data they compiled: Average Employment per fast food restaurant (Before and after the date of the minimum wage increase in NJ) NJ PA Before (March 92) 20.44 23.33 After (Dec 92) 21.03 21.17 QUESTION 1 : Can we interpret the change in employment over time in NJ (after vs. before, i.e., 21.03 – 20.44 = 0.59) as a result of a higher minimum wage? QUESTION 2 : Can we interpret the difference in employment in December 1992 between NJ and PA (i.e., 21.03 – 21.17 = -0.14) as the result of a higher minimum wage?
Background image of page 2
Empirical Methods II (API-202) Kennedy School of Government Harvard University 3 QUESTION 3 : How can we use PA to mimic the counterfactual for NJ, controlling for long- term differences between NJ and PA? ANSWER: With the difference-in-differences estimator (diff-in-diff, DID) : 1 DID = ( EMP 2 NJ – EMP 1 NJ ) - ( EMP 2 PA – EMP 1 PA ) (1:before 2:after) = (21.03-20.44) - (21.17-23.33) = 2.75 2 The DID estimator allow us to control for: o Long term differences in employment between NJ and PA – ie, differences between the states that existed before and after. In practice:
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/12/2009 for the course HKS API202A taught by Professor Levy during the Spring '09 term at Harvard.

Page1 / 9

LN+11+Diff+in+Diff - Empirical Methods II (API-202) Kennedy...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online