This is the third of three blogs co-written by Con Keating and Iain Clacher informing on the current consultation on the DB funding code of practice,
The other articles can be accessed at the bottom of this blog
Optimal Scheme Management
For the calculation of scheme liabilities this should be done as either the accrued value of contributions, or the discounted present value of the projected (unbiased i.e. with no prudence) best estimate of the ultimate benefits using the contractual accrual rate (CAR). While for scheme funding, we advocate that this should be to the level of liabilities calculated in this manner, with assets being valued using market prices. However, this leaves the scheme facing just one risk, sponsor insolvency at a time when this level of funding is insufficient to purchase equivalent benefits in the open market or alternately, to run off the scheme. There is a solution to this risk – pension indemnity assurance; a specialised form of long term credit insurance.
This class of business is effectively, offered by the PPF but in a manner that can be considered as both deficient and inefficient. The deficiency arises from the offer of only partial benefits; there is no valid justification for these cuts to member benefits. The inefficiency can be seen in the PPF’s excessive levy charges – the evidence for which is the massive reserves they have accumulated in a very short period, and this levy is effectively paid with corporate capital. This therefore raises the question: what is the likely cost of pension indemnity assurance for schemes funded to the level of best estimate? Here the PPF provides a poor guide, given its history of excessive levy charges.
There are many variables which determine the premium in any case. These include the quality of the sponsor covenant, the degree of maturity of the scheme together with its status e.g. whether it is open or closed, and the basis of calculation of the best estimate. Dependent upon these specifics, modelling indicates that the premium will lie in the range 0.2%-0.5% of liabilities for all but a handful of companies that are in particularly bad shape. It is also worth stating this again – this is insurance of the full benefits, not PPF reduced variants. The interval for premium setting may be either fixed for scheme life or set at some shorter period, such as annually.
Business models for writing pension indemnity assurance
There are two classes of business model possible. These differ in the annuity and deferred annuity treatment post insolvency; the indemnity assurer may either purchase these from another insurer/reinsurer at the time of sponsor insolvency or it may write these itself and run off the pension liabilities. Purchasing annuities on insolvency is less efficient from a return on capital standpoint as it reduces the modelled return on capital by between 15 and 25%.
This business model is also not hypothetical, with the most notable specialty insurer of this class of business being PRI-Pensionsgaranti. Several specialty insurers have also offered a related instrument, surety bonds, though these have a fixed pay-off rather being benefits related. PRI-Pensionsgaranti has already written a small number of policies in the UK, mainly in a form qualifying as contingent assets. In Sweden it has written insurance for some 1400 entirely unfunded, book-reserve schemes. It operates the less efficient business model of purchasing annuities from another company on an insolvency event. It charges 0.4% of liabilities as a premium and has been highly profitable. As a mutual, it rebates these as dividends to the member schemes. This premium is not a pure sunk cost; unlike the PPF arrangement, the policy is also an asset of the scheme and not a contingent asset. It is also interesting to note that as in this asset role it is an automatic stabiliser for the scheme.
PR-Pensionsgaranti has even insured funded schemes specifically to allow them to move from funded to unfunded status. This liberates scheme assets for corporate business investment. There are complications with respect to the taxation of such liberations, but the Swedish treatment has been at the generous end of the spectrum; no taxation if reinvested in corporate activities within a specified, short period. If liabilities were to be calculated using the CAR best estimate, many schemes would be materially overfunded and so the liberation of some scheme assets may be both feasible and desirable especially at a time when corporate profits and cash flow are likely to be severely constrained.
Advantages of pensions indemnity insurance for scheme management
One major advantage of the use of pension indemnity insurance is that the duties placed on trustees are significantly reduced. Trustees would be concerned only with the administration of pensions and the current valuation of collateral assets. No integrated risk management, no long-term objective. Consideration of the future performance of assets and all the risk management practices associated with that would be redundant and as such would save significant sums of money for the scheme and its sponsor.
For the sponsoring employer, they would be concerned with the extent to which the assets held as collateral security would defray their future costs. As such, the asset management style they adopt would be a matter of their risk tolerance and would most likely follow a simple risk and return analysis. Liability driven investment would not be necessary as the hedging of the spuriously introduced interest rate sensitivities is no longer required or desirable. As these currently account for around 40% of overall scheme assets in many larger pension funds, the net returns to sponsor employers should, ceteris paribus, be significantly higher than those which have or might be achieved under the existing or proposed models.
Creating a market for pensions indemnity insurance
To ensure a competitive market for pension indemnity assurance, private sector provision should be permitted and encouraged. The PPF should be required to offer full benefits cover and should also be privatised as an industry owned mutual insurance company. Shares in this mutual should be awarded free of cost to schemes in proportion to the total levy contributions they have made. This would augment overall scheme assets by an amount of the order of £6 billion and generate approximately £2 billion in tax revenues.
If the funding to best estimate model is followed, it will reduce significantly, almost entirely, the requirement for special contributions, reducing the tax lost under those arrangements by £2-£3 billion per year. The total cost of pension indemnity assurance premiums would be approximately £3 billion if all schemes adopted it, and the tax cost would be of the order of £750 million. Of course, in this situation there would be no justification for schemes participating in the PPF scheme as is or paying the £550 million of the scheme levy.
If funding were to be to the CAR best estimate, and all sponsors were to liberate their excess assets, the windfall tax gain would be of the order of £45 billion at standard corporation tax rates, and in the range £90-£100 billion at 55%, though at this latter rate it is highly unlikely that many sponsors would wish to liberate. The amounts available to be freed for new corporate investment would lie in the range £150-£250billion.
The DB funding consultation
The proposed new scheme funding requirements carry with them an annual tax cost of at least £1 billion, rising to £3 billion after a few years, and perhaps considerably more. The Regulator has stated: “Our funding consultation is about the balance between member security and affordability.”, which makes the absence of any cost/benefit analysis in the 175 pages of the consultation document noteworthy.
The consultation is principally concerned with management of the ’end-game’ for closed schemes now running off, but open ongoing schemes, though few in number, are not treated differently. The traditional solution of bulk annuitisation has been supplemented by a wide range of proposals and products; some, such as the efforts of the consolidators, have been widely publicised. others, such as surety bonds and Legal and General’s “insured self-sufficiency” less so.
In this series of blogs, we have concentrated on a few of the problems arising from DB pension regulations and the practices they induce. Our coverage was deliberately minimalist, there are many other issues which we might have highlighted. It is true that some people suffered and lost pensions in the pre-Goode days, but they were far fewer in number than the many millions who received their pensions routinely over many decades. The ‘fixing of pensions’ has been completely disproportionate and its cost outrageous – running to hundreds of billions of pounds and the remedy has killed all too many schemes which were in rude health.
Pensions need a bonfire of regulation this blog calls for a radical shake-up of the rules governing DB pensions
The following blogs are designed to be read consecutively.
The other way to value DB schemes
A different approach to pension scheme solvency and funding
The best way to manage a DB scheme
“For the calculation of scheme liabilities this should be done as either the accrued value of contributions, or the discounted present value of the projected (unbiased i.e. with no prudence) best estimate of the ultimate benefits using the contractual accrual rate (CAR). While for scheme funding, we advocate that this should be to the level of liabilities calculated in this manner, with assets being valued using market prices. However, this leaves the scheme facing just one risk, sponsor insolvency at a time when this level of funding is insufficient to purchase equivalent benefits in the open market or alternately, to run off the scheme. There is a solution to this risk – pension indemnity assurance; a specialised form of long term credit insurance.”
I would think that the British Steel Pension Scheme (BSPS2) has ‘pension indemnity assurance’ in case of sponsor insolvency?
“the indemnity assurer may either purchase these from another insurer/reinsurer at the time of sponsor insolvency or it may write these itself and run off the pension liabilities. Purchasing annuities on insolvency is less efficient from a return on capital standpoint as it reduces the modelled return on capital by between 15 and 25%.”
1. Would it be possible to spell out the explanation for the reduction in modelled returns? Does this have to do with application of solvency requirements to annuity provides (such as 99.5% confidence and Minimum Capital Requirement) which would not apply to your preferred approach to indemnity insurance?
2. Would the assets of your indemnity insurer be invested much more heavily in ‘return-seeking’ equities and property as opposed to bonds than the portfolios of insurance companies that provide annuities or the portfolio of the PPF?
3. If the answer to 2 is ‘yes’, how would you respond to the objection that the value of these return-seeking assets would be vulnerable to fall in price during an economic crisis which gives rise to extensive pension sponsor insolvencies, and that the insurer therefore needs to countercyclically invest the proceeds of the levy in bonds rather than equities and property?
(Apologies for mistakes in the above that reflect my ignorance of how insurance is regulated.)