Prof Mike Otsuka on First Actuarial and USS’ different approaches to self-sufficiency!

This article has been published by Mike Osuka on his blog. You can read the original here; http://tinyurl.com/j82en9s .

prof

Professor Mike Otsuka

 

Mike is a  Professor in the Department of Philosophy, Logic & Scientific Method at the London School of Economics. He’s kindly permitted his blog to be republished here


The UK Universities Superannuation Scheme (USS) has recently released its ‘Proposed Approach to the Methodology for the 2017 Actuarial Valuation’. The Universities and Colleges Union (UCU) have commissioned First Actuarial, a consultancy held in high regard by pensions professionals, to provide an analysis. Two of their actuaries, Hilary Salt and Derek Benstead, have now written a report for UCU.

Their report challenges the role and nature of USS’s self-sufficiency valuation, which assumes investment primarily in gilts (government bonds), and argues for its replacement by a self-sufficiency valuation that assumes investment primarily in equity (stocks and shares).

‘Self-sufficiency’ plays a crucial role for USS because it figures in their much-discussed Test 1. Here’s how USS characterises this test:

“Test 1: This test measures the reliance being placed on the covenant. Specifically it measures the difference between the technical provisions and the amount of assets which would be required for self-sufficiency…. At the 2014 valuation the trustee decided that the reliance should not exceed the value of additional contributions (above the agreed 18% of salary) that they believed could be available from the participating employers over a 15–20 year period. Advice received by the trustee at that time indicated 25% of salary was an upper limit (i.e. a further 7%).” (‘Proposed Approach’, p. 13)

The ‘technical provisions’ corresponds to the ‘prudent’ discount rate that is used to determine whether the ongoing pension scheme is fully funded at the triennial valuations and to set pensions contributions.

The ‘covenant’ is the employer’s ability to provide financial support to the pension scheme, via additional contributions.

USS’s rationale for Test 1 is apparently that, over the visible period up to a time horizon past which it is no longer possible to assess it, the covenant is strong enough for the employer to increase their contributions from the current 18% of salaries to 25% of salaries, but no higher. Beyond such an increase in employer contributions, pensions obligations would need to be paid almost entirely out of assets in the scheme. In other words, the scheme would have to become ‘self-sufficient’ beyond this point.

What level of assets, and of what kind, are required for self-sufficiency? This is what USS says:

‘[a] The ‘self-sufficiency’ measure of the liability reflects the required level of assets to meet all future benefit payments to a very high probability without the need for additional contributions. [b] It corresponds to a discount rate very close to gilts (i.e. very close to the yield on an appropriate portfolio of UK government bonds). The discount rate used by the trustee for the self-sufficiency liability is gilts + 0.5%.’ (‘Proposed Approach’, p. 10, n. 5, my bold lettering added)

Although Salt and Benstead don’t describe it as such, I think it helpful to understand First Actuarial’s response as involving the acceptance of [a] while denying [b]. In other words, First Actuarial denies that USS would need to be invested primarily in gilts in order to almost guarantee that it can meet pensions obligations out of nothing other than assets and the associated investment income.

First Actuarial’s argument is that, given everything we now know about long term returns on equity versus gilts, it would be far less expensive to invest nearly everything in equity rather than gilts in order to achieve self-sufficiency as defined by [a].

The level of investment in equity that First Actuarial maintains would be enough to nearly guarantee the payment in full of pensions obligations is based on historical analysis. They take the return on equity for every period between 1920 and 2015, understood as the return from 1920 to 2015, from 1921 to 2015, from 1922 to 2015, etc. The level that is sufficient to nearly guarantee full payment is that which would have fully covered pensions obligations, even during the worst performing of these historical time intervals.

This argument is captured in the following graph on p. 15 of First Actuarial’s response.

As I have argued in this linked document (see the Appendix on p. 9 entitled ‘USS’s Test 1 chain to the ball of sinking gilt yields is the biggest problem with the valuation methodology’), self-sufficiency understood as gilts + 0.5% serves as a ball and chain for the following reason: the more the gilt yield sinks, the more USS is forced by Test 1 to de-risk its assets out of higher return equity and into lower-return bonds. It is forced to engage in such de-risking in order to ensure that an increase from 18% to 25% in the employer contribution rate is enough to finance the transformation of their actual mixture of higher-return assets into a self-sufficiency portfolio heavily invested in sinkingly-lower-return gilts which nevertheless generates the same amount of income.”

If, however, First Actuarial is right, USS is mistaken in assuming that, pound for pound, investment in gilts provides a more effective means than investment in equity of meeting pensions promises almost for certain. Therefore, Test 1, which is driving a costly de-risking of the assets of the pension scheme out of equity and into gilts, rests on a mistake.

prof-2

Mike Otsuka in the LSE


A postscript to my ‘First Actuarial’s response’ post

I can see the following rationale for USS’s approach:

Even if one thinks there’s no chance that the UK higher education sector will otherwise go bust, there is still a limit beyond which employers cannot increase their pensions contributions (measured in terms of % of salary), without cutting so much into their ability to spend money on other things that they are no longer able to support teaching and research properly.

The covenant is just that limit.

According to the latest review, this covenant is visible up to a 30 year horizon.

Even if one thinks universities will surely be around in 50 and 100 years, it might be difficult to assess what this limit will be beyond 30 years. Perhaps beyond 30 years, even a 25% employer contribution would be unsustainable, given economic and political conditions. So a 30 year horizon is not implausible.

If, moreover, there were some non-negligible probability of having to increase pensions contributions up to this limit (i.e., from 18% to 25%) during the 30 year visibility of the covenant, then I can see the rationale for a requirement that it be possible, without breaching this limit, to transform the pension scheme into one that is self-sufficient.

Even if we accept all of the above (which I am inclined to accept), we still have the following two good responses to USS:

1. Equity would deliver self-sufficiency more effectively than gilts. (That’s one of the main points of the First Actuarial paper, and the one I focus on in my blog post to which this is a postscript.)

2. It’s only because of USS’s unrealistically volatile, mark-to-market, gilts-pegged valuation methodology that there is a decent probability that so great a funding shortfall will arise at a future valuation that an increase from 18% to 25% would be necessary. On First Actuarial’s Internal Rate of Return (IRR) approach to the valuation, with its more realistic focus on the annual cash flows that will be available over time to pay pensions as they come due, there is a much lower chance that employer contributions would have to increase by that much.

This second point has been emphasised in previous documents from First Actuarial. It is also the focus of this earlier blog post of mine.

About henry tapper

Founder of the Pension PlayPen, Director of First Actuarial, partner of Stella, father of Olly . I am the Pension Plowman
This entry was posted in actuaries, pensions and tagged , , , , , , , . Bookmark the permalink.

5 Responses to Prof Mike Otsuka on First Actuarial and USS’ different approaches to self-sufficiency!

  1. Andrew Main says:

    Henry
    As far as I understand it, and I am by far and expert but from numbers I have seen most University pension funds are “advised” to be only 77% vested of what is needed. If they are only 77% vested how does this effect the return calculations assumed. Have they been taken into account by both groups looking at the actuarial assumptions? Are they storing a problem for tomorrow especially when returns maybe way below the assumptions.

    Like

  2. Mike Otsuka says:

    Thanks, Henry, for cross-posting this! I have now posted the following postscript to my above post:

    View story at Medium.com

    I’ve also edited the second-to-last paragraph of the above post to make things clearer. That paragraph now reads as follows:

    “…self-sufficiency understood as gilts + 0.5% serves as a ball and chain for the following reason: the more the gilt yield sinks, the more USS is forced by Test 1 to de-risk its assets out of higher return equity and into lower-return bonds. It is forced to engage in such de-risking in order to ensure that an increase from 18% to 25% in the employer contribution rate is enough to finance the transformation of their actual mixture of higher-return assets into a self-sufficiency portfolio heavily invested in sinkingly-lower-return gilts which nevertheless generates the same amount of income.”

    Liked by 1 person

  3. Why use UK equities for this purpose? If you’re trying to solve for the confidence level at which equities match the gilt yield (surely an ILG yield?), and you choose to rely on a mean-reverting equity real-return model (because it’s evident in all markets not just the UK), along with a currency model in which real exchange rate risk (ie deviation from PPP with mean = zero) is allowed for, surely you would want to take advantage of the risk-reducing effects of global diversification?
    Our modelling of sterling-adjusted real equity returns to match a stream of long-term liabilities suggests 99% confidence of exceeding ILGs for almost all cash flows longer than about 10 years – that’s way less than ‘normal’ and entirely thanks to the level of interest rates. There’s nothing odd about the equity markets.
    As you’ve pointed out Henry, it’s the same logic for DB transfers.

    Liked by 1 person

  4. Mike Otsuka says:

    I’ve just posted a further follow-up to the above post:

    View story at Medium.com

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s