Experiments and Natural Experiments

1 748 659 585 550 immigrant grew up with two parents

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: ssignment(%) Months on IA last 3 years IA continuouslylast 3 years (%) Percenton IA by months since randomassignment Month 6 90.8 Month 12 83.7 Month 18 77.9 Month 24 73.0 Month 36 65.4 Month 48 56.7 Month 60 50.6 Month 69 45.0 Numberof observations 2,786 2,831 1,874 aSample includes observations in the SSP Recipient Experiment who were on IA in the month of random assignment and the previous month. Eligible program group is the subset who received at least one SSP subsidy payment. D. CARD AND D. R. HYSLOP 1734 and control groups. Unfortunately, these data have some critical limitations based Income Assistance data. Most imporrelative to the administratively tantly, they are only available for 52 months after random assignment.Since some programgroup memberswere still receivingsubsidypaymentsas late as month 52, this time window is too short to assess the long-run effects of the program.Indeed, looking at Figure la, there is still an impact on IA participation in month 52 that does not fully dissipate until month 69. Second, because of nonresponses and refusals,labor marketinformationis only available for 85% of the experimental sample (4,757 people).l8 Third, there appear to be relativelylarge recall errors and seam biases in the earnings and wage data.19 Nevertheless, the labor market outcomes provide a valuable complement to based welfare participationdata. the administratively 3 shows the average monthly employment rates of the programand Figure control groups, along with the associated experimental impacts. After random assignment the employment rate of the control group shows a steady Source: Card and Hyslop (2005) 0.5 - 0.4 . .. 0.3 -- Control ~ Group =~~c;^^^^~ ? 0.2 -----*- v?C;1~~~~~~~~ Program Group Difference 0.1 0 6 12 18 24 30 36 42 48 Months Since Random Assignment FIGURE3.-Monthly employment rates. 18The distribution of response patterns to the 18-, 36-, and 54-month surveys is fairly simi- lar for the programand control groups (chi-squaredstatistic= 11.4 with 7 degrees of freedom, p-value = 0.12). However, a slightlylarger fraction of the programgroup have complete labor market data for 52 months-85.4% versus 84.0% for the controls. Moreover, the difference in mean IA participationbetween the treatmentand control groupsin month 52 is a little different in the overallsample (2.5%) than in the subsetwith complete labormarkethistories(3.3%). of "9Each the three post-random-assignment surveysasked people about their labor market outcomes in the 18 months since the previoussurvey.Manypeople reportconstantearningsover the recall period, leading to a pattern of measured pay increases that are concentratedat the seams, ratherthan occurringmore smoothlyover the recall period. 1738 D. CARD AND D. R. HYSLOP The absence of a trend in the averagemarginalwage relativeto the minimum wage is importantbecause it suggeststhat the SSP programgroup experienced little or no relative gain in potential wages over the course of the experiment. This is confirmedby the analysisin TableIII of labor market outcomes in the last availablemonth (month 52). Recognizingthe higheraveragelevel of wages in...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online