“…In addition the means-tested benefit for pensioners, known as the Minimum Income Guarantee (MIG), was formally indexed to earnings rather than prices from April 1999 (unlike the rest of the pension programme), so increasing its real value for low earners and reinforcing the disincentive to save for retirement. Analysis of these, and subsequent, trends suggests that, at least under the current regime, whereas replacement rates cohort-by-cohort for the public pension programme have already peaked for average earners, low earners are likely to see increasingly generous replacement rates from the public programme for several decades yet (Disney and Emmerson, 2005). To the extent that single women, or couples that include a zero earner or have low education (to take some of the groups where we have found a significant effect) are disproportionately likely to gain from these reforms to the public pension programme relative to the control group, we might expect to understate take-up as a result of the change in contribution limits.…”