The rise in contributions traces to an inexplicable refusal to allow a surplus

USS’s response to my concerns over Test 1

In mid-October, I posted a blog, which has now been viewed nearly 10,000 times, in which I argued that Sam Marsh’s modelling of asset projections had revealed that “USS’s valuation rests on a large and demonstrable mistake”. A few days later USS posted extended responses here and here, to which I replied in a blog post entitled “Test 1 de-risking is dead, long live de-risking!”.

In late November, USS’s Chief Risk Officer Guy Coughlan sent me the following linked further response entitled “USS briefing on claims of an error in the 2017 valuation”, which he has also given me permission to publicly post.

USS’s latest response raises more questions and concerns than it answers. It opens as follows:

The claim: “USS erroneously assumed its assets plus target reliance matched its liabilities at year 20 instead of checking what its own projections actually showed. Once this is corrected, there is no need to de-risk, reduce benefits or increase contribution rates.”

Under legislation, assets equalling liabilities beyond the recovery plan period isn’t an erroneous assumption, it’s a given: we know that we will have to adjust our assets (via contributions and investment strategy at every valuation) to equal [MO: See point ‘a’ below] the Technical Provisions liabilities by that point. Further, the trustee’s Statement of Funding Principles, which are agreed with participating employers, clearly states that the trustee’s aim is to hold assets equal [MO: See point ‘b’ below] to the Technical Provisions.

Test 1 was designed with this in mind, and consciously sets the Technical Provisions liabilities at 2037 as self-sufficiency less target reliance — not in error but because, as above, the assets will be constructed to equal [MO: See point ‘c’ below] the Technical Provisions at this time.

I emailed Coughlan to ask for clarification regarding the following:

a. Shouldn’t this be “at least equal”? My understanding is that current legislation and regulation do not rule out the scheme’s going into surplus. There is, for example, no longer government pressure related to tax avoidance which stands in the way of the scheme’s being in surplus.

b. I can’t find anywhere in the SFP where this aim of assets “equal”, as opposed to “at least equal”, to technical provisions liabilities is stated. Would it be possible to cite the relevant passage you have in mind? Of course, in order to fulfil the SFO, the trustee will aim for assets at least equal to TP liabilities. But that’s consistent with a surplus. Moreover, running a substantial surplus on the current return-seeking portfolio would appear to be a cost-effective means of simultaneously meeting the following two aims to which USS is committed: (1) fully recovering from any TP deficit, and (2) ensuring that the reliance on the covenant is no greater than £10 bn in 20 years’ time. (See my comments below c.)

c. Again, I think this should be “at least equal”.

As Sam Marsh has shown with his asset projections that USS has confirmed, if the scheme remains invested in the current return-seeking portfolio, rather than de-risking, and carries on with 26% contributions, there is a prudent 67% chance that the scheme will be in substantial surplus in 20 years’ time. In fact, this surplus will be so substantial that the reliance on the covenant (i.e., the gap between actual assets and the assets required to purchase a self-sufficiency portfolio) will be less than the currently specified £10 bn.

Hence, (i) on USS’s current understanding of reliance on the covenant in terms of asset gap, (ii) which is the only understanding that makes sense of the underlying rationale for Test 1 (see this blog post for a defence of both (i) and (ii)), there is no need to de-risk the portfolio in order to remain within the currently specified £10 bn reliance gap in 20 years’ time.

An important upshot is that one does not even need to accept JEP’s proposed weakening of the 20 year target reliance from £10 bn to £13 bn in order to reject any Test 1 based case for a de-risking of the portfolio over 20 years. No such de-risking is required on a proper understanding of Test 1, given its underlying rationale, even if target reliance remains at £10 bn.

In saying this, I grant that USS may have other grounds for de-risking the portfolio or raising contributions, which you go on to spell out in the remainder of your reply to me, where these grounds have to do with such things as short-term reliance, downside risk, gilt yield reversion, or what the regulator will find acceptable.

Any such multi-faceted approach to risk-management should, however, take on board the possibility of effectively managing risk by means of running a TP surplus on a return-seeking portfolio. It appears, however, from what you say in the opening paragraphs of your reply to me that this possibility has been excluded, without obvious justification.

Any light you can shed on why you say “equal” rather than “at least equal” would be greatly appreciated.

Coughlan replied to confirm that he did in fact mean “equal” rather than “at least equal”. He drew attention to the passage in which he writes that “we will have to adjust our assets (via contributions and investment strategy at every valuation) to equal the Technical Provisions liabilities”, while also stating that the trustee would neither allow a significant surplus nor a significant deficit to arise.

What was lacking, however, was an explanation for the trustee’s alleged refusal to allow a significant surplus.

There is, moreover, the following compelling reason for them to allow such a surplus to arise: it would place the scheme within affordable £10 bn reach of USS’s much touted safe harbour of self-sufficiency.

As an alternative to allowing a surplus to arise, USS has adopted a policy of de-risking the assets to the point where the scheme would be within the required £10 bn gap to self-sufficiency even if it is merely fully funded rather than in surplus. To achieve full funding by means of such a de-risked portfolio, contributions would eventually need to increase by 10.6% overall, to 24.9% for employers and 11.7% for members, under the 2017 Rule 76 valuation that USS will soon be submitting to the regulator.

In short, USS’s case for de-risking the assets, at great cost of increase in contributions, is based on an unexplained and inexplicable refusal to allow a significant surplus, in spite of the compelling reason they have to do so: namely, that it would place the scheme within affordable distance of the safe harbour of self-sufficiency.

Before USS launch their consultation on a new valuation that calls for a rise in the current rate of contributions, scheme members, employers, and trustees are owed an explanation from Coughlan and other USS executives of their refusal to allow such a surplus. Such an explanation is owed, especially given the similar concerns First Actuarial has raised in the section entitled ‘Reliance, discount rate and investment strategy’ on p. 4 of this linked paper.

Update 05 April 2020: Please see this Twitter thread — plus those embedded in the thread — which provides an update to my above exchange with USS: