Why it makes no sense, under the dual discount rate, for any USS employer to advocate 100% DC
[See also this sequal to this post: ‘Why Trinity’s departure and the abandonment of Test 1 render a shift to 100% DC problematic’]
The case for closing DB and moving to 100% DC is undermined by the dual discount rate (DDR) which JEP2 recommended, the “vast majority” of employers endorsed, and USS has adopted. Here’s why:
Before DDR, there was the following case against closure of DB to future accrual: as First Actuarial has often pointed out, that would quickly render the scheme cash flow negative, thereby giving rise to disinvestment risk, as one would be forced to sell assets to pay pensions as they fall due, even if market conditions tell against selling.
The above case against DB closure remains. But the DDR creates the following additional reason to keep DB open:
Under the DDR, the pre-retirement years of member pensions promises are discounted at the higher rate of growth assets than their post-retirement years, which are discounted at the lower rate of a bond-weighted self-sufficiency portfolio.
Moreover, under the DDR, each year of future accrual is discounted at a higher rate than the sum total of past accrual. This is because each year of future accrual will consist of a much greater proportion of pre-retirement years than post-retirement years, in comparison with the sum total of past accrual which is valued at the same point in time.
As a result, when the scheme remains open with a membership profile that remains roughly constant over time, the proportion of pre-retirement (growth asset) years to post-retirement (self-sufficiency) years will remain fairly constant at roughly the current proportions of 55% growth assets, 45% self-sufficiency.
This in spite of the fact that past accrual — i.e., accrual built up over the decades leading up to 31 March 2020 valuation date — will become increasingly weighted towards a self-sufficiency portfolio as members who were promised pensions in past years move from active membership to retirement.
This higher discount rate for future accrual under an open scheme provides a rationale for the assumption that the investment of deficit recovery contributions (DRCs) in future years will outperform the discount rate on past accrual. As JNC member Sam Marsh notes in this tweet:
With an open scheme, one can justify such outperformance by noting that one is discounting at no greater a rate than the rate for future DB accrual, while also noting that such future accrual keeps the scheme cash flow positive, thereby eliminating disinvestment risk.
With a closed scheme, one lacks the above grounds for an outperformance assumption, since there is no longer a higher discount rate of future DB accrual to which to anchor one’s outperformance assumption, and there is no longer the safety from disinvestment risk of the positive cash flow to which such future accrual gives rise. Rather, all we will have is past accrual, which will become increasingly mature and weighted towards a self-sufficiency portfolio, as more of those to whom these DB pensions have been promised up to date of scheme closure move from active membership to retirement over time.
This is, moreover, a feature of the DDR, not a bug. The DDR was introduced to facilitate the joint UUK-UCU commitment to find a way to make it affordable to continue to provide DB for future accrual.
It flies in the face of the DDR for employers to push for 100% DC. Those ‘hawkish’ employers who would like to close DB and replace it with 100% DC will need to tear up the recommendations of JEP2 and repudiate UUK’s overwhelming past commitment to DDR.
Under the current valuation framework which employers have embraced, future accrual of DB in some significant form is the only game in town, and employers had better figure out how to make this work, rather than pushing for 100% DC.
The red-underlined passage in this Universities UK employer consultation document implies that UUK has received indication from USS that a shift to 100% DC would place upward pressure on DRCs, since they would hope for a small change only if they expect any changes to be upward rather than downward: