Experiments and Natural Experiments

G imbens et al use alternative control groups not as

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: e standard errors 2) Hours elasticity of 0.6 3) Total elasticity (unconditional hours) is 0:4+0:6 = 1 Caveats: 1) Does the common trends assumption hold? Potential story biasing the result: Trend toward “power couples" and thus DD might not be due to taxes: In 1983-1985, high income husbands had non-working spouses, In 1989-1991, high income husbands married to professionals [and no change for middle class] 2) LFP before the reform is very different across T and C groups ) DD sensitive to functional form assumption [such as levels vs logs] 3) Liebman and Saez (2006) plot full time-series CPS plot and show that Eissa's results are not robust using admin data (SSA matched to SIPP) Fortin – Econ 560 Lecture 1C In a follow-up study, Eissa and Liebman (1996) consider the labour force participation of single mothers using CPS data o used a difference-in-differences strategy o to examine the relatively small 1986 EITC expansion in the US o comparing single mothers (treatment) to single women without children (control) o in the periods 1984-1986 (before) vs. 1988-1990 (after) Pr(lfpit 1) ( Z it 0 treat i 1 post 86 t 2 (treat post 86) it ), They find a small but significant DD effect: 2.4% (larger DD effect 4% among women with low education) o Translates into substantial participation elasticities above 0.5 o Conventional standard errors probably overstate precision o LFP for women with/without children are not great comparison groups (70% LFP vs. +90%): time series evidence is only moderately convincing Subsequent studies have used much bigger EITC expansions of the mid 1990s and also find positive effects on labour force participation of single women/single mothers As we saw previously, earning income tax credit generates a piece-wise linear budget constraint with kinks. Source: Eissa and Hoynes (2008) Figure 8 Stylized EITC Budget Constraint After Tax Income D Phase-out w(1-τp) C Flat w B No EITC w Phase-in w(1+τs) A Hours of Work 0 38 Source: Eissa and Liebman (1996), p. 631 Source: Eissa and Liebman (1996) z 0 >z ) e- D 00 | 0 ; O 0 0 t q 6 O O 0 ~ 6 0 0 O O O0O 0 0 0 O O 0 Cz 00 O -4 0 O -4 O 0 o o o 6 cq 0 0 0~~~~~C O '-i C0 '- C> CS - 0 s O O 0 0 c o O 0 oto O 66. 0 C O 0 o O C C) 0 0 O C O 0 .0 A l 44 := 0 C)~~~~~~~~~0 O 6 0 0 0 O 0 CO C) ~ ~~~~~~~~~~~o ~ ~~ ~~~ CO 10 t m O 0 > ; H 0 c 0~~~~~~~~~~~~~~~~~~~~ 0 ' o0 Ot- 0 C oo X 0 0 O 0 > 0 'm ow 0~ 4 0 O cO 0 0 '- ~~ 10 H :H^A '-4 0 U 0 | I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~o a'booz . | . 0. || O 00 0 0 0~~~~~~~~~~~~~~~~0 OO z 04' o 0-' C O~~~~~~~~~~O 0 0~~~~~ 0 oc o oU 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 00 E* - l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 X 4~~~~~~~~~~~~~~ 0 0 0? 0. 0 Q~~~~ 0 0 Source: Blundell and Hoynes (2004) Has “In-Work”Benefit Reform Helped the Labor Market? 3000 441 .0.0500 2500 E m t n -0.1000 0 v 3 g 2000 d El L 0 m i P) 5 Maximum EITC Difference 1500 .0.15002 E“ W c g I .c - , ---., 1000 I ‘ .-’ E n ,’ Employment Rate Difference .0.2000 I 500 0 .0.2500 1984 1986 1988 1990 1992 1994 1996 1998 Year Fig. 10.17 Maximum EITC and difference in annual employment rates (comparison...
View Full Document

This document was uploaded on 02/26/2014 for the course ECON 560 at UBC.

Ask a homework question - tutors are online