401(k) participation by pay band, for employees hired before automatic enrolment and under it

Each line names a claim related to this exhibit, its source, and how it was checked.

• A MSH01-C2 Participation: the WINDOW cohort (hired April 1997 to March 1998, not automatically enrolled) and the NEW cohort (hired April 1998 to March 1999, automatically enrolled) compared at 3 to 15 months of tenure on 30 June 1998 and 30 June 1999: 37.4 and 85.9 per cent (Table IV, p. 1160, page image; N = 4,249 and 5,801); raw difference 48.5 points, 50.4 regression-adjusted for gender, race, age, compensation and months of tenure, both marked significant (Table VI, p. 1167); immediate eligibility, introduced the same day, checked on the OLD cohort (30 June 1998) and the WINDOW cohort (30 June 1999) at 15 to 27 months: 48.7 and 49.4 per cent, 0.6 raw and 4.1 adjusted, the adjusted estimate also marked significant (Table VI, page image; G5 N1), small relative to the combined impact, the comparison not completely untainted; Table IV by compensation (page image), WINDOW then NEW: under 20 thousand dollars 12.5 and 79.5; 20 to 29 thousand 24.5 and 82.8; 30 to 39 thousand 42.2 and 88.9; 40 to 49 thousand 51.0 and 91.8; 50 to 59 thousand 61.6 and 92.8; 60 to 69 thousand 59.7 and 94.7; 70 to 79 thousand 57.9 and 91.5; 80 thousand and over 68.3 and 94.2; the highest and lowest pay-related rates about 55 points apart (WINDOW) and 15 (NEW); effects largest among the groups with the lowest participation before; participation equalized, in the conclusion partially (p. 1185; G5 N12); the authors attribute the difference almost completely to automatic enrollment; a comparison of hiring cohorts, not of employees assigned at random (pp. 1151, 1154 to 1162, 1167, 1185): "were made immediately eligible to participate in the 401(k) plan regardless of service"; "less than half the 86 percent participation rate of the NEW cohort with a similar amount of tenure"; "The regression-adjusted impact of automatic enrollment on the 401(k) participation rate for these two groups of employees is in fact even slightly larger, at 50 percent"; "The regression-adjusted differences are estimated from OLS regressions which include categorical controls for gender, race, age, compensation, and months of tenure."; "An asterisk indicates statistical significance."; "This is small, however, relative to the overall combined impact of automatic enrollment and immediate eligibility"; "This comparison does not provide a completely untainted assessment of the impact of immediate eligibility"; "the disparity between the highest and lowest pay-related participation rates is about 55 percentage points for the WINDOW cohort"; "15 percentage points for the NEW cohort"; "The effects are largest among the groups with the lowest participation rate under the previous regime of affirmative elections"; "it equalizes participation rates across various demographic groups"; "Thus, it seems likely that the 401(k) participation differences between the NEW and the WINDOW cohort can be attributed almost completely to the impact of automatic enrollment."; "Automatic enrollment also serves to partially equalize participation differences"Madrian & Shea (2001) · Madrian & Shea. (2001)

Grades: A, verified against the printed page of the primary source · B, primary source, text layer only · C, authoritative secondary · D, reported.

This exhibit belongs to an essay: the numbers live here, the argument lives there.