PensionReforms' summary and comments
In the US (and elsewhere) private Defined Benefit, Tier 3 retirement savings schemes are in decline.  Defined Contribution schemes are taking their place.  In part, that change seems to be driven by changes to the reporting standards for DB schemes in 2006 - see here.  But there are other influences at work in the US because the same kind of thing is happening in other countries.

The report records that DB membership among private sector employers in the US fell from 38% of employees to 20% between 1980 and 2008.  Over the same period, membership of Defined Contribution schemes (where that is the only scheme) increased from 8% to 31%.  Some think that most private sector DB schemes in the US will be first frozen and then terminated.

"This paper uses the Model of Income in the Near Term ["MINT"] to simulate the impact of an accelerated transition from DB to DC pensions on the distribution of retirement income among boomers.  A scenario in which employers freeze all remaining private sector DB plans and a third of all state and local plans over the next five years will on balance produce more losers than winners among boomers and reduce their average incomes at age 67."

In fact, the report says that "...the losers greatly outnumber the winners."

PensionReforms notes that this conclusion looks just at the benefits produced by the changed mix of Tier 3 schemes.  That there will probably be more losers than winners means that the subsidies paid by employers to Tier 3 schemes may be less; alternatively, direct, immediate remuneration to employees is higher (or a mix of these).  In the first case (lower subsidies), employers' profits might be higher and shareholders, perhaps, better off (or consumers if the employers' prices are lower as a consequence); in the second case (higher remuneration), whether employees are in fact worse off depends on what they do with the extra pay.  In other words, it depends where public policy considerations 'rule off' the comparison.

Putting a specific estimate on the impact:
"While most boomers will experience modest changes in income, an accelerated decline in DB coverage will reduce incomes by at least 5 percent for about 10 percent of last wave boomers."

The report suggests that, based on its modelling, the impact on eventual benefits will be distributed unevenly:
"Income changes will be largest among higher-income boomers, who have the highest DB coverage rates and projected pension incomes.  Furthermore, the numbers of winners and losers and net income changes are much greater for the last wave of boomers (born between 1961 and 1965) than for earlier boomers.  Younger boomers are most likely to have their DB pensions frozen with relatively little job tenure and to lose their high accrual years for DB pension wealth, but also to have relatively more years to accumulate DC pension wealth before retirement."

PensionReforms notes that the impact on the younger baby boomers is a function of time to retirement and the increased probability of encountering a frozen DB scheme or no DB scheme at all.  The impact on the more highly paid reflects the regressive nature of both the Defined Benefit promises themselves and the value of the tax concessions that go with them.

As the report notes, actual "[i]ndividual outcomes will depend on the magnitude of DB pension losses, participation and contribution rates in the new DC pension plans, and investment returns on retirement account assets."

PensionReforms thinks that, from a public policy perspective, the change from DB to DC is probably a positive move in the long term.  It will mean more transparent remuneration arrangements that will be better understood by shareholders, regulators, tax collectors and the employees themselves.  It will also reduce the tax breaks attributable to DB schemes that advantaged the higher paid and so were regressive.  DC schemes are also regressive (the higher paid can better afford to join and collect the maximum subsidies available) but are likely to be less generous and more open about it. (File size 208 KB; 49 pp) 348