Table of Contents |
Figure ES.1: Estimated Eligibility for and Use of the ERA Stipend Among 100 TANF Applicants and Recipients in Corpus Christi
This flow chart illustrates the eligibility criteria for and the use of the ERA stipend for all ERA group members randomly assigned from October 2000 through January 2003 in Corpus Christi. This chart shows the experiences of 100 typical TANF applicants and recipients who are randomly assigned to ERA. Eighty four of those randomly assigned received TANF during the follow-up period and of those 84, 70 were employed during the follow-up period and received TANF. Next, 55 of these 70 applicants and recipients were employed for four months within a year during the follow-up period. Finally, 30 of the 55 received at least one stipend.
Figure 2.1: Typical Paths of Individuals Through the Texas ERA Program
This flow chart illustrates the typical paths of individuals through the Texas ERA program toward job readiness and job search services and/or retention and advancement services. After completing their TANF application or redetermination interview, individuals are randomly assigned to the ERA program and attend the TANF workforce orientation. If the individuals are not approved for TANF, they are not eligible for TANF or ERA. If individuals are approved for TANF, they can start to work on job readiness and job search services. If individuals do not become employed after these activities, they participate in unpaid volunteer or community service positions. Once individuals are employed, they receive retention and advancement services. After 12 stipends, individuals exit the ERA program.
Figure 2.2: Percentage of Program Group Members Receiving the ERA Stipend in Corpus Christi, by Quarter After Random Assignment
This line graph shows both the cumulative percentage of program members receiving the ERA stipend in Corpus Christi, and also the percentage first receiving stipend, by the quarter after random assignment.
The x axis displays the quarter of first stipend from Quarters 1 through 16.
The y axis displays the percentage of sample members on a scale of 0 to 50.
Over the 16 quarters, the cumulative percentage receiving the stipend increased fairly steadily. In Quarter 1 the cumulative percentage of sample members receiving a stipend is 0.3 percent. This percentage increases to 3.4 in Quarter 2 and 9.0 in Quarter 3. The cumulative percentage continues to increase until it reaches 32.7 percent in Quarter 16.
The percentage of sample members first receiving the stipend starts out at 0.3 percent during Quarter 1. The percentage then increases to 3.1 in Quarter 2 and it peaks to 5.6 in Quarter 3, and then decreases gradually to less than one percent in Quarter 16.
Figure 2.3: Percentage Receiving the ERA Stipend among the Full Sample and Key Subgroups Defined by Pre-Random Assignment Status
This bar graph illustrates the percentage receiving the ERA stipend among the full sample and key subgroups defined by pre-random assignment status. The x axis depicts percentage points on a scale of 0 to 45, and the y axis lists the names of five subgroups.
The top of the bar graph illustrates that among the full sample, 20.4 percent of sample members in Houston received the stipend, while 21.7 percent of sample members received it in Fort Worth, and 30.0 percent in Corpus Christi.
In Houston, 25.4 percent of sample members employed in the prior year (according to UI records) received a stipend. In Fort Worth, 27.2 percent of the same subgroup received a stipend and 35.1 percent of this subgroup received the stipend in Corpus Christi.
For sample members not employed in the prior year (according to UI records), the following percentages received the ERA stipend: 12.6 percent in Houston, 16.4 percent in Fort Worth, and 26.3 percent in Corpus Christi.
For the subgroup of sample members that earned greater than $5,000 in the prior year (according to UI records), 28.8 percent received a stipend in Houston, 30.4 percent in Fort Worth, and 40.0 percent in Corpus Christi.
Finally, the last three bars on the bottom of the bar graph display the percentage of sample members who earned less than $5,000 in the prior year (according to UI records) and received the ERA stipend. In Houston, 23.2 percent of this subgroup received the stipend, and in Fort Worth and Corpus Christi, 24.7 and 32.0 percent received the stipend respectively.
Figure 2.4:Estimated Eligibility for and Use of the ERA Stipend Among 100 TANF Applicants and Recipients in Corpus Christi
This flow chart illustrates the eligibility criteria for and the use of the ERA stipend for all ERA group members randomly assigned from October 2000 through January 2003 in Corpus Christi. This chart shows the experiences of 100 typical TANF applicants and recipients who are randomly assigned to ERA. Eighty four of those randomly assigned received TANF during the follow-up period and of those 84, 70 were employed during the follow-up period and received TANF. Next, 55 of these 70 applicants and recipients were employed for four months within a year during the follow-up period. Finally, 30 of the 55 received at least one stipend.
Figure 2.5: Summary of How Texas ERA Case Managers Typically Spend Their Time
These three pie charts show how Texas ERA case managers typically spend their time in each site. The first pie chart shows that in Corpus Christi, case managers spent a total of 22 percent of their time working with clients, of which 10 percent was spent in contact with working clients and 11 percent was spent in contact with non-working clients. They spent 78.5 percent of their time working on other activities which included 13 percent working on administrative duties, 6 percent attending staff meetings, 14 percent outreaching to clients, 14 percent completing counselor notes, 8 percent interacting with employers, 5 percent traveling to program activities or meeting with clients or employers, and 19 percent on other miscellaneous activities.
The second pie chart shows that in Houston, case managers spent a total of 29 percent of their time working with clients, of which 22 percent was spent in contact with working clients and 12 percent was spent in contact with non-working clients. They spent 71 percent of their time working on other activities which included 10 percent working on administrative duties, 10 percent attending staff meetings, 9 percent outreaching to clients, 9 percent completing counselor notes, 7 percent working on TANF participation, 7 percent working on job development, 6 percent traveling to program activities or meeting with clients or employers, 4 percent monitoring client participation in services, and 4 percent on other miscellaneous activities.
The third pie chart shows that in Fort Worth, case managers spent a total of 28 percent of their time working with clients, of which 12 percent was spent in contact with working clients and 16 percent in contact with non-working clients. They spent 71 percent of their time working on other activities which included 16 percent working on administrative duties, 7 percent attending staff meetings, 11 percent observing or assisting with specific program activities, 9 percent completing counselor notes, 6 percent monitoring client participation in services, and 23 percent on other miscellaneous activities.
Figure 4.1: Percentage Employed According to UI Records and Receiving the ERA Stipend, Corpus Christi
Figure 4.1 shows the percentage who were employed and receiving the stipend and the percentage who were employed without receiving the stipend in each quarter of the follow-up period in Corpus Christi. The control group’s bars show the employment rates that would have been expected in the absence of ERA. To the extent that the shaded portion of the ERA group’s bar overlaps with the control group’s bar, the stipend was paid to those who would have worked anyway. To the extent that the shaded region exceeds the control group’s employment bar, the stipends encouraged new employment. Although the ERA group’s bars are mostly a little higher than the control group’s, the figure shows that stipends do not appear to have added much additional employment.
For example, in Quarter 1, 49 percent of the ERA Group and 51 percent of the Control Group was employed and not receiving the ERA stipend. In Quarter 2, 52.1 percent of the ERA Group was employed, including 3.1 percent who were employed and were receiving the stipend and 49 percent who were not receiving the stipend. Forty nine percent of the Control Group was employed during Quarter 2, and none of these sample members were eligible for the ERA stipend.
Quarters 5, 6, and 7 have the highest percentage of ERA sample members employed and receiving the stipend. For example, in Quarter 5, 12.6 percent of the ERA group was employed and receiving the stipend and 41 percent of the ERA group was employed and not receiving the stipend. In Quarters 6 and 7, 14.2 and 13.7 percent of the ERA group was employed and receiving the ERA stipend, respectively. In Quarter 6, 39 percent of the ERA group was employed and not receiving the stipend, and in Quarter 7, 40 percent of the ERA group was employed and not receiving the stipend. Fifty percent of the Control Group was employed in Quarter 5. In Quarters 6 and 7, 51 and 53 percent of the Control Group was employed respectively (and not receiving the stipend).
Appendix Figure E.1: Impacts on TANF Payments and TANF Receipt Rates for the Report Sample and Early Cohort Over Time
This exhibit displays two line graphs for each of the three ERA sites in Texas: Corpus Christi, Fort Worth, and Houston. Three of the six graphs show the impact of the percentage receiving TANF in any given quarter in both the report sample and early cohort, and the remaining three graphs show impact of TANF payments per quarter for both the report sample and early cohort.
In all of the graphs showing the impacts on TANF receipt rates the x axis displays the quarter relative to random assignment, for Quarters 1 through 11. The y axis shows the percentage receiving TANF in any given quarter, on a scale of -20 percent to 20 percent, in increments of 5 percent. One line on the graph represents the impacts for the early cohort (for Quarters 1 through 11), and another line represents the impacts of TANF receipt for the report sample (for Quarters 1 through 7). In all of the graphs showing the impacts on TANF payments, the x axis displays the quarter relative to random assignment, for Quarters 1 through 11, and the y axis shows the TANF payments, ranging from -$200 to $200, in increments of $50. One line on the graph represents the impacts for the early cohort (for Quarters 1 through 11), and another line represents the impacts of TANF receipt for the report sample (for Quarters 1 through 7).
In Corpus Christi the early cohort was randomly assigned from October 2000 to June 2001, and the report sample was randomly assigned from October 2000 to June 2002. The figure showing impacts on TANF receipt rates shows that overall there were not large impacts on the percentage of sample members receiving TANF in any given quarter. The impacts for both the early cohort and report sample increase from Quarter 1 to Quarter 2, and then they mostly decrease over time. Differences were statistically significant at the 10 percent level for the early cohort in Quarter 2. For the report sample, differences were statistically significant at the 5 percent level in Quarter 6 and the 10 percent level in Quarter 7.
The figure that displays impacts on TANF payments in Corpus Christi also shows that there were no large impacts, and no impacts were statistically significant. The impact of TANF payments peaks for both the early cohort and report sample in Quarter 2 at $19.50 and $18.20 respectively. The report sample decreases to -$15.00 in Quarter 7 and the early cohort decreases to -$20.80 in Quarter 9 and then increases to -$5.60 in Quarter 11.
In Fort Worth the early cohort was randomly assigned from October 2000 to June 2001, and the report sample was randomly assigned from October 2000 to June 2002. The figure showing impacts on TANF receipt rates shows that overall there were not large impacts on the percentage of sample members receiving TANF in any given quarter, and no impacts were statistically significant. The impact for the early cohort was the greatest in Quarter 3, where it reached 3.8 percent. It was smallest in Quarter 4, where it was -2.1 percent. The impact was the greatest for the report sample in Quarter 3, where it was 0.5 percent, and it was the smallest in Quarter 4, where it was -3.1 percent.
The figure that displays impacts on TANF payments in Fort Worth also shows that there were no large impacts, and no impacts were statistically significant. The impact of TANF payments for the early cohort never goes above $21.20 nor below -$17.70, and the impact for the report sample never goes above $6.00 nor below -$17.60.
In Houston the early cohort was randomly assigned from March to June 2001, and the report sample was randomly assigned from March 2001 to June 2002. The figure showing impacts on TANF receipt rates shows that overall there were not large impacts on the percentage of sample members receiving TANF in any given quarter. The impacts for the early cohort increase from Quarter 2 to Quarter 5, decrease from Quarters 6 to 8, increase in Quarter 9, and then gradually decrease in Quarters 10 and 11. Differences were statistically significant at the 10 percent level for the early cohort in Quarters 5 and 6. For the report sample, the impacts stayed relatively the same over time, and the differences were statistically significant at the 10 percent level in Quarter 1.
The figure that displays impacts on TANF payments in Houston also shows that there were no large impacts. The impact of TANF payments is statistically significant at the 10 percent level for the early cohort in Quarters 3 and 6, and at the 5 percent level in Quarter 9. The impact for the report sample is statistically significant at the 10 percent level in Quarter 5.
Appendix Figure E.2: Estimated Eligibility for and Use of the ERA Stipend Among 100 TANF Applicants and Recipients in Fort Worth
This flow chart illustrates the eligibility criteria for and the use of the ERA stipend for all ERA group members randomly assigned from October 2000 through January 2003 in Fort Worth. This chart shows the experiences of 100 typical TANF applicants and recipients who are randomly assigned to ERA. Eighty three of those randomly assigned received TANF during the follow-up period and of those 83, 68 were employed during the follow-up period and received TANF. Next, 52 of these 68 applicants and recipients were employed for four months within a year during the follow-up period. Finally, 22 of the 52 received at least one stipend.
Appendix Figure E.3: Estimated Eligibility for and Use of the ERA Stipend Among 100 TANF Applicants and Recipients in Houston
This flow chart illustrates the eligibility criteria for and the use of the ERA stipend for all ERA group members randomly assigned from October 2000 through January 2003 in Houston. This chart shows the experiences of 100 typical TANF applicants and recipients who are randomly assigned to ERA. Eighty nine of those randomly assigned received TANF during the follow-up period and of those 89, 68 were employed during the follow-up period and received TANF. Next, 49 of these 68 applicants and recipients were employed for four months within a year during the follow-up period. Finally, 20 of the 49 received at least one stipend.
Appendix Figure E.4: Impacts on Program Participation for Key Subgroups in Corpus Christi
This bar chart displays the impact on program participation for various subgroups in Corpus Christi. The x axis displays percentage points on a scale of 0 to 25, in increments of 5, while the y axis displays four different subgroups of sample members.
The first three bars on the chart show the impacts on program participation for the subgroup of 49 sample members who earned $5,000 or more in the prior year. ERA increased the percentage receiving help with retention/advancement by 7.2 percentage points and it increased the percentage receiving help with job preparation by 5.6 percentage points. The program also increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 22.3 percentage points.
The next cluster of bars on the chart displays impacts on program participation for the 116 sample members who earned less than $5,000 in the prior year. ERA increased the percentage receiving help with retention/advancement by 13.7 percentage points and it increased the percentage receiving help with job preparation by 11.2 percentage points. The program also increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 8.7 percentage points.
The next three bars on the chart show the impacts on program participation for the 77 sample members who were unemployed in the prior year. The ERA program increased the percentage receiving help with retention/advancement by 15.2 percentage points. ERA had no impact on the percentage receiving help with job preparation. The program increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 11.9 percentage points.
The last cluster of three bars on the figure displays impacts on program participation for the 213 sample members who were employed in the prior year. ERA increased the percentage receiving help with retention/advancement by 13.4 percentage points, and this impact was statistically significant at the 5 percent level. The ERA program increased the percentage receiving help with job preparation by 1.4 percentage points. The program also increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 15.6 percentage points, an impact that was statistically significant at the 5 percent level.
Appendix Figure E.5: Impacts on Program Participation for Key Subgroups in Fort Worth
This bar chart displays the impact on program participation for various subgroups in Fort Worth. The x axis displays percentage points on a scale of -40 to 40, in increments of 10, while the y axis displays four different subgroups of sample members.
The first three bars on the chart show the impacts on program participation for the subgroup of 42 sample members who earned $5,000 or more in the prior year. ERA decreased the percentage receiving help with retention/advancement by 14.9 percentage points and it decreased the percentage receiving help with job preparation by 28.7 percentage points. The program increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 5.1 percentage points.
The next cluster of bars on the chart displays impacts on program participation for the 70 sample members who earned less than $5,000 in the prior year. ERA increased the percentage receiving help with retention/advancement by 30.0 percentage points, which is statistically significant at the 5 percent level. ERA increased the percentage receiving help with job preparation by 24.5 percentage points, and it also increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 24.5 percentage points. Both of these impacts are statistically significant at the 10 percent level.
The next three bars on the chart show the impacts on program participation for the 49 sample members who were unemployed in the prior year. The ERA program increased the percentage receiving help with retention/advancement by 9.3 percentage points. ERA increased the percentage receiving help with job preparation by 3.6 percentage points. The program decreased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 4.3 percentage points.
The last cluster of three bars on the figure displays impacts on program participation for the 139 sample members who were employed in the prior year. ERA increased the percentage receiving help with retention/advancement by 18.4 percentage points, and this impact was statistically significant at the 5 percent level. The ERA program increased the percentage receiving help with job preparation by 5.5 percentage points. The program also increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 20.8 percentage points, an impact that was statistically significant at the 5 percent level.
Appendix Figure E.6: Impacts on Program Participation for Key Subgroups in Houston
This bar chart displays the impact on program participation for various subgroups in Houston. The x axis displays percentage points on a scale of -20 to 50, in increments of 10, while the y axis displays four different subgroups of sample members.
The first three bars on the chart show the impacts on program participation for the subgroup of 45 sample members who earned $5,000 or more in the prior year. ERA increased the percentage receiving help with retention/advancement by 16.3 percentage points and it increased the percentage receiving help with job preparation by 47.1 percentage points. The program increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 36.6 percentage points.
The next cluster of bars on the chart displays impacts on program participation for the 135 sample members who earned less than $5,000 in the prior year. ERA increased the percentage receiving help with retention/advancement by 5.6 percentage points and it decreased the percentage receiving help with job preparation by 2.8 percentage points. ERA also decreased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 7.7 percentage points.
The next three bars on the chart show the impacts on program participation for the 78 sample members who were unemployed in the prior year. The ERA program increased the percentage receiving help with retention/advancement by 7.3 percentage points. ERA increased the percentage receiving help with job preparation by 7.8 percentage points. The program also increased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 19.6 percentage points.
The last cluster of three bars on the figure displays impacts on program participation for the 219 sample members who were employed in the prior year. ERA increased the percentage receiving help with retention/advancement by 6.2 percentage points. The ERA program increased the percentage receiving help with job preparation by 0.4 percentage points. The program decreased the percentage of sample members that had any contacts with a case manager/employment program since random assignment by 2.5 percentage points.
Appendix Figure E.7: Percentage Employed According to UI Records and Receiving the ERA Stipend, Fort Worth
Appendix Figure E.7 shows the percentage who were employed and receiving the stipend and the percentage who were employed without receiving the stipend in each quarter of the follow-up period in Fort Worth. The control group’s bars show the employment rates that would have been expected in the absence of ERA. To the extent that the shaded portion of the ERA group’s bar overlaps with the control group’s bar, the stipend was paid to those who would have worked anyway. To the extent that the shaded region exceeds the control group’s employment bar, the stipends encouraged new employment. Although the ERA group’s bars are mostly a little higher than the control group’s, the figure shows that stipends do not appear to have added much additional employment.
For example, in Quarter 1, 46 percent of the ERA Group and 44 percent of the Control Group was employed and not receiving the ERA stipend. In Quarter 2, 47.3 percent of the ERA Group was employed, and of this percentage employed, 2.3 percent were receiving the stipend and 45 percent were not receiving the stipend. Forty six percent of the Control Group was employed during Quarter 2, and none of these sample members were eligible for the ERA stipend.
Quarters 4, 5, 6, and 7 have the highest percentage of ERA sample members employed and receiving the stipend. For example, in Quarter 4, 9.0 percent of the ERA group was employed and receiving the stipend and 40 percent of the ERA group was employed and not receiving the stipend. In Quarters 5, 6 and 7, 9.8, 11.5, and 12.3 percent of the ERA group was employed and receiving the ERA stipend respectively. In Quarter 5, 38 percent of the ERA group was employed and not receiving the stipend, and in Quarter 6, 38 percent of the ERA group was employed and not receiving the stipend. In Quarter 7, 36 percent of the ERA group was employed and not receiving the stipend. Forty seven percent of the Control Group was employed in Quarter 4, 48 percent in Quarter 5, 46 percent in Quarter 6, and 47 percent in Quarter 7.
Appendix Figure E.8: Percentage Employed According to UI Records and Receiving the ERA Stipend, Houston
Appendix Figure E.8 shows the percentage who were employed and receiving the stipend and the percentage who were employed without receiving the stipend in each quarter of the follow-up period in Houston. The control group’s bars show the employment rates that would have been expected in the absence of ERA. To the extent that the shaded portion of the ERA group’s bar overlaps with the control group’s bar, the stipend was paid to those who would have worked anyway. To the extent that the shaded region exceeds the control group’s employment bar, the stipends encouraged new employment. Although the ERA group’s bars are mostly a little higher than the control group’s, the figure shows that stipends do not appear to have added much additional employment.
For example, in Quarter 3, 3.3 percent of the ERA Group was employed and receiving the stipend. Forty percent of the ERA group was employed and not receiving the ERA stipend and 44 percent of the Control Group was employed and not receiving it.
Quarters 6, 7, and 8 have the highest percentage of ERA sample members employed and receiving the stipend. For example, in Quarter 6, 7.2 percent of the ERA group was employed and receiving the stipend and 37 percent of the ERA group was employed and not receiving the stipend. In Quarters 7 and 8, 8.0 and 7.9 percent of the ERA group was employed and receiving the ERA stipend respectively. In Quarter 6, 37 percent of the ERA group was employed and not receiving the stipend, and in Quarter 7, 36 percent of the ERA group was employed and not receiving the stipend. In Quarter 8, 34 percent of the ERA group was employed and not receiving the ERA stipend. In Quarters 6, 7, and 8, 41 percent of the Control Group was employed in each quarter.
Appendix Figure E.9: Impacts on UI-Covered Earnings and Employment for the Report Sample and Early Cohort Over Time
This exhibit displays two line graphs for each of the three ERA sites in Texas: Corpus Christi, Fort Worth, and Houston. Three of the six graphs show the impact of ERA on the percentage employed in both the report sample and early cohort, and the remaining three graphs show the impact of ERA on quarterly earnings for both the report sample and early cohort.
In all of the graphs showing the impact of ERA on the percentage employed the x axis displays the quarter relative to random assignment, for Quarters 1 through 13. The y axis shows the impact of ERA on the percentage employed, on a scale of -20 percent to 20 percent, in increments of 5 percent. One line on the graph represents the impacts for the early cohort (for Quarters 1 through 13), and another line represents the impact of ERA on the percentage employed for the report sample (for Quarters 1 through 9). In all of the graphs showing the impact of ERA on quarterly earnings, the x axis displays the quarter relative to random assignment, for Quarters 1 through 13, and the y axis shows impact of ERA on quarterly earnings, ranging from -$500 to $500, in increments of $100. One line on the graph represents the impacts for the early cohort (for Quarters 1 through 13), and another line represents the impacts for the report sample (for Quarters 1 through 9).
In Corpus Christi the early cohort was randomly assigned from October 2000 to June 2001, and the report sample was randomly assigned from October 2000 to June 2002. The line for the early cohort on the figure showing impacts of ERA on the percentage employed starts at 2.1 percent in Quarter 1. The line increases and decreases until Quarter 8, at which point it gradually increases to 9.8 percent. The impacts for the early cohort are statistically significant at the 10 percent level for Quarters 2, 9, and 11. The impacts of ERA on the percentage employed are statistically significant at the 5 percent level for the early cohort in Quarters 5, 10, and 12. The impact in Quarter 13 is statistically significant at the 1 percent level. For the report sample, the impacts mostly increase from Quarters 1 through 5, decrease to Quarter 7, and then increase to Quarter 9. The impact of ERA on the percentage employed for the report sample is the highest in Quarter 9, where the impact is statistically significant at the 1 percent level.
The figure that displays impacts of ERA on quarterly earnings in Corpus Christi shows that the line representing the early cohort increases from Quarters 1 through 4, mostly declines until Quarter 7 and then starts to mainly increase until Quarter 12, and then decreases in Quarter 13. Impacts are statistically significant for the early cohort in Quarters 3 and 12 at the 10 percent level and in Quarter 4 at the 5 percent level. The impact of ERA on quarterly earnings for the report sample peaks in Quarter 4 and then decreases until Quarter 8. The impact in Quarter 4 is statistically significant at the 10 percent level.
In Fort Worth the early cohort was randomly assigned from October 2000 to June 2001, and the report sample was randomly assigned from October 2000 to June 2002. The line for the early cohort on the figure showing impacts of ERA on the percentage employed increases from Quarters 1 to 3, decreases until Quarter 5, goes up and down from Quarters 6 through 10 and then increases until Quarter 12. The impact for the early cohort is statistically significant at the 10 percent level in Quarter 3. None of the impacts for the report sample are statistically significant and they generally stay the same except for a dip in Quarter 5 and a peak in Quarter 6.
The figure that displays impacts of ERA on quarterly earnings in Fort Worth shows that the line representing the early cohort is the lowest in Quarters 5 though 7 and the highest in Quarter 12, with no impacts being statistically significant. The line for the report sample shows that impacts were the highest in Quarter 9 (at $240); this impact is statistically significant at the 5 percent level. The impacts were the lowest for the report sample in Quarters 2, 5, and 7.
In Houston the early cohort was randomly assigned from March to June 2001, and the report sample was randomly assigned from March 2001 to June 2002. The impact of ERA on the percentage employed in the early cohort generally increases from Quarters 2 through 6, decreases until Quarter 10, and then increases until Quarter 13. None of the impacts for the early cohort are statistically significant. The line for the report sample is the lowest in Quarter 3 and it increases until Quarter 6, and then decreases until Quarter 9. In Quarter 6, the impact is statistically significant at the 10 percent level for the report sample.
The figure that displays impacts of ERA on quarterly earnings in Houston shows that the line representing the early cohort is the lowest in Quarter 3 and the highest in Quarter 13. The line representing the report sample is the highest at Quarter 6 and lowest in Quarter 9. No impacts on this figure are statistically significant.
| Table of Contents |