New Evidence on Earnings and Benefit Claims Following Changes in the Retirement Earnings Test in 2000

by Jae G. Song and Joyce Manchester
ORES Working Paper No. 107 (released June 2006)
Table equivalent for Chart 1. Transition from not working to working, by age at the end of t1
Transition Age 65 Age 66 Age 67 Age 68 Age 69
t1=1996 t2=1997 2.30 2.22 2.21 1.76 1.61
t1=1997 t2=1998 2.17 2.38 2.05 1.67 1.68
t1=1998 t2=1999 2.39 2.13 2.09 1.74 1.94
t1=1999 t2=2000 2.87 3.05 2.52 2.52 2.18
t1=2000 t2=2001 2.05 1.90 1.80 1.61 1.79
t1=2001 t2=2002 1.90 1.93 1.83 1.65 1.36
t1=2002 t2=2003 2.08 2.07 2.02 1.71 1.45
SOURCE: Authors' tabulations using the 1 percent extract of the Social Security Administration's Master Earnings File and Master Beneficiary Record.
Table equivalent for Chart 2. Transition from not entitled to entitled, by age at the end of t1
Transition Age 65 Age 66 Age 67 Age 68 Age 69
t1=1996 t2=1997 1.76 0.97 0.72 0.65 1.24
t1=1997 t2=1998 1.96 1.06 0.69 0.67 1.49
t1=1998 t2=1999 2.62 1.20 1.06 0.74 1.52
t1=1999 t2=2000 4.72 2.96 2.00 1.61 1.74
t1=2000 t2=2001 0.75 0.59 0.48 0.50 1.09
t1=2001 t2=2002 0.86 0.37 0.27 0.42 0.87
t1=2002 t2=2003 1.32 0.85 0.35 0.41 1.12
SOURCE: Authors' tabulations using the 1 percent extract of the Social Security Administration's Master Earnings File and Master Beneficiary Record.
Table equivalent for Chart 3. Probability of an increase in earnings if earnings at t1 are greater than or equal to zero or greater than zero, by age at t1
Transition Age 65 Age 66 Age 67 Age 68 Age 69
Earnings at t1 greater than or equal to zero
t1=1996 t2=1997 16.58 14.75 13.30 11.10 9.97
t1=1997 t2=1998 17.36 15.91 13.89 12.01 10.41
t1=1998 t2=1999 17.73 16.20 14.10 11.99 11.01
t1=1999 t2=2000 19.44 18.10 16.23 13.70 11.57
t1=2000 t2=2001 19.35 17.74 15.56 13.86 11.69
t1=2001 t2=2002 18.37 17.08 14.57 13.46 11.16
t1=2002 t2=2003 18.40 16.32 14.95 13.01 11.19
Earnings at t1 greater than zero
t1=1996 t2=1997 14.38 12.68 11.27 9.52 8.56
t1=1997 t2=1998 15.26 13.66 12.01 10.51 8.93
t1=1998 t2=1999 15.41 14.17 12.16 10.42 9.29
t1=1999 t2=2000 16.65 15.18 13.85 11.39 9.64
t1=2000 t2=2001 17.36 15.92 13.86 12.36 10.09
t1=2001 t2=2002 16.53 15.22 12.84 11.93 9.92
t1=2002 t2=2003 16.38 14.33 13.04 11.42 9.87
SOURCE: Authors' tabulations using the 1 percent extract of the Social Security Administration's Master Earnings File and Master Beneficiary Record.
NOTE: Earnings are in current dollars.

Text description for Chart 4.
Distribution of old-age beneficiaries with earnings in $1,000 intervals relative to the earnings test threshold for treatment and control groups, before and after removal

This bar graph consists of four panels. Two panels are labeled younger groups, one of these shows data for 1996–1999 and the other shows data for 2000–2003. The other two panels are labeled older groups, likewise with one panel for 1996–1999 and the other for 2000–2003.

The vertical axis in all four panels is percent; in the younger groups panels it ranges from zero to 4.0 in increments of 0.5, and in the older groups panels ranges from zero to 1.2 in increments of 0.2

In all panels, the horizontal axes represent $1,000 earnings intervals relative to the earnings test threshold; five intervals below the threshold and four intervals above the threshold are shown. Because thresholds vary from year to year, horizontal axes are not labeled with specific values.

The panels for the younger groups show data separately for individuals turning ages 62–64 and those turning age 65 during the period of coverage.

For those turning ages 62–64 in 1996–1999, the percentage in each earnings interval increases gradually from about 2.0 percent to about 2.5 percent between the fifth and the second interval under the threshold, then spikes at about 3.8 percent in the first interval under the threshold. In each successive interval above the threshold, the percentages decline steadily from about 2.1 percent to about 0.9 percent. The threshold values for this group are $8,280, $8,640, $9,120, and $9,600.

For those turning ages 62–64 in 2000–2003, the percentage in each earnings interval increases gradually from about 1.9 percent to about 2.2 percent between the fifth and the second interval under the threshold, then spikes at about 3.1 percent in the first interval under the threshold. In each successive interval above the threshold, the percentages decline steadily from about 1.8 percent to about 0.9 percent. The threshold values for this group are $10,080, $10,680, $11,280, and $11,520.

For those turning age 65 in 1996–1999, the percentage in each earnings interval fluctuates, from about 1.5 percent in the fifth interval under the threshold to about 1.9 percent in the first interval under the threshold. In each successive interval above the threshold, the percentages decline slightly from about 1.5 percent to about 1.3 percent. The threshold values for this group are $8,280, $8,640, $9,120, and $9,600.

For those turning age 65 in 2000 2003, the percentage in each earnings interval decreases slightly throughout all nine intervals, ranging from about 0.8 percent in the lowest earnings interval to about 0.4 percent in the highest earning interval. The threshold values for this group are $17,000, $25,000, $30,000, and $30,720.

The panels for the older groups show data separately for individuals who have attained ages 65–69 and those who have attained ages 70–72 during the period of coverage.

For those who have attained ages 65–69 in 1996–1999, the percentage in each earnings interval fluctuates slightly, from a minimum of just below 1.0 percent in the second interval under the threshold to a maximum of about 1.1 percent in the first interval under the threshold. In each successive interval above the threshold, the percentages decline steadily from about 0.6 percent to about 0.3 percent. The threshold values for this group are $12,500, $13,500, $14,500, and $15,500.

For those who have attained ages 65–69 in 2000–2003, the percentage in each earnings interval gradually declines across eight of the nine intervals, ranging from about 0.8 percent in the lowest earnings interval to about 0.4 percent in the highest earning interval. The threshold values for this group are $16,500, $17,500, $18,500, and $19,500.

For those who have attained ages 70–72 in 1996–1999, the percentage in each earnings interval gradually declines across eight of the nine intervals, ranging from just below 0.6 percent in the lowest two earnings intervals to just above 0.2 percent in the highest earning interval. The threshold values for this group are $12,500, $13,500, $14,500, and $15,500.

For those who have attained ages 70–72 in 2000–2003, the percentage in each earnings interval decreases slightly throughout all nine intervals, ranging from about 0.5 percent in the lowest earnings interval to about 0.2 percent in the highest earning interval. The threshold values for this group are $16,500, $17,500, $18,500, and $19,500.

To summarize, the percentages of individuals in each earning interval generally decline from the lowest earning interval to the highest earning interval. The notable exceptions are those turning ages 62–64 in 1996–1999, and to a lesser extent those turning ages 62–64 in 2000–2003; for these groups, the percentage in each earnings interval increases gradually between the fifth and the second interval under the threshold, spikes at the first interval under the threshold, then declines steadily in each successive interval above the threshold.

Table equivalent for Chart 5. Estimates of the effects on earnings, by percentile and year
Percentile 1997 1998 1999 2000 2001 2002 2003
Have attained ages 65–69
10th 0.00056 0.03683 -0.02053 0.02908 -0.07619 -0.00565 -0.11714
20th -0.20244 -0.29371 -0.34971 -0.32041 -0.53094 -0.44509 -0.54072
30th -0.16369 -0.36904 -0.47323 -0.67247 -0.63701 -0.74058 -0.9167
40th -0.34477 -0.52449 -0.59893 -0.5643 -0.6323 -0.69133 -0.92513
50th -0.31006 -0.65198 -0.67478 -0.48731 -0.24625 -0.28436 -0.34423
60th -0.32557 -0.367871 -0.31126 0.13524 0.46723 0.80307 0.869911
70th -0.258689 -0.61414 -0.81221 -0.23893 0.533749 1.03955 1.24946
80th -0.218482 -0.755865 -0.99323 -0.65697 0.746954 0.985766 1.06506
90th -0.158158 -1.30889 -1.83359 -2.23557 -1.32472 -1.42702 -1.45545
Turning age 65
10th 0.17485 0.20848 0.2435 0.19588 0.31106 0.18193 0.26088
20th 0.194055 0.384275 0.23257 0.410235 0.339365 0.396805 0.47337
30th 0.27262 0.29078 0.08286 0.317939 0.081929 0.323979 0.36899
40th 0.176199 0.2872 0.127579 0.99647 0.484259 0.744749 0.75802
50th 0.051597 0.041761 -0.27601 1.532 1.44591 2.27664 2.03508
60th 0.049531 0.547431 -0.54001 1.70617 1.75988 2.53831 2.39915
70th -0.250603 0.589625 -1.3417 1.27845 1.15906 1.62569 2.54424
80th -0.779865 -0.099789 -1.45721 0.711451 -0.22342 0.035698 0.415678
90th -1.59195 -0.275619 -1.87307 0.257538 -1.05071 -0.53049 0.260109
SOURCE: Authors' estimates.
NOTE: The dependent variable is annual earnings in thousands of dollars. The samples include observations with nonzero earnings. Other covariates used in this regression are a constant, male, race (white), age group dummies (62–64 and 70–72), and calendar-year dummies from 1997 through 2002.