Don’t stop believing is a sequel to the October 2011 paper Don’t stop thinking about tomorrow, which considers in detail the position of UK private occupational pensions. It is thought-provoking: to quote one reviewer: “In his two “Don’t Stop” papers Con Keating has “watched” pensions more closely than others and uncovered the grand narrative. The first paper considered pensions in general terms, providing critical arguments absent from the debate. The second – this paper – considers the specifics of the UK economic position and proposes explicit remedies. Both are a “must read” for policy makers, whether considering state pensions as a means of social welfare or occupational pensions as a means of deferred pay.
The first conclusion is that collective provision triumphs over individual provision. So the Don’t Stop papers are a must read for employers too.”
It analyses a number of commonly-held beliefs and finds them unsubstantiated by evidence:
1) Pensions are unaffordable and unsustainable
The reality is that pensions are both affordable and sustainable in the public and private sector. Those Jeremiahs who point to an increasing share of our national output as the cost of pensions exaggerate the position. The fact is that true pensions costs are rising less rapidly than economic output. They also fail to recognise that as our output and wealth grows so we will want to spend proportionately more on education, healthcare and retirement.
2) The individual should provide for his own pension
It is true that an individual should save towards her pension, but that does not mean that the institutional design of pensions saving should be individual in nature. These arrangements inevitably lead to extreme dependencies on financial market performance and are highly inefficient. For the same pension result, individual DC will cost at least 50% more in contributions than collective DB.
3) Pensions costs have risen dramatically because of increasing longevity
Pensions costs have risen, but at a rate significantly lower than national output and far lower than private sector earnings. True pensions costs have doubled over the past two decades while the funding cost of pensions has more than quadrupled. This is a result of ill-conceived regulation operating on top of accounting standards which are not fit for purpose and badly mis-state the health of schemes.
4) Employers should not be in the business of providing pensions
A pension is a claim of future production – the output of employer sponsors. Employers can credibly contract with employees on their future production, without the inter-position of financial markets. In fact, the prevalent analysis of pensions in terms of financial analysis, when this is at heart a labour market contract is misconceived. These pension contracts, though different in detail, are in essence the same as company-issued equity or debt which is traded in financial markets.
5) The accounting standards are merely the messenger of the parlous state of DB pensions.
This is simply untrue. The accounting standards in force, and the currently proposed amendments to them, understate the true strength of the position and introduce spurious volatility to scheme financial statements and those of their sponsor employer. This has real and very costly effects.
The accounting standards take a conventional financial balance sheet view when the inter-temporal nature of the pension contract makes the income and expense approach appropriate. The correct discount rate to apply to the discounting of pension liabilities is the company’s rate of return on capital employed, and for unfunded public sector pensions, the rate of GDP growth.
6) Pensioners are now better protected by UK pension regulations
Regulation has raised the cost of DB pension provision to the point that it is uneconomic for employer sponsors to provide pensions rather than cash wages. Apart from the time inconsistency of rules based upon the funding levels of schemes, there is a paucity of evidence that members have been better protected. More importantly, the most significant cost of these regulations is the number of employees who no longer benefit from occupational DB schemes.
One of the most inept pieces of institutional design is the Pension Protection Fund. This is a mutual compensation fund; it is not an insurance company. The consequence of this is that all of the levies paid to it and the costs of compliance, which approach £1 billion annually, are sunk costs and pure expense to schemes. Insurance, by contrast, would be an asset of the scheme.
7) Asset and liability management techniques can resolve the problem.
Pension fund asset allocations have changed dramatically and management techniques have become much more sophisticated. Notwithstanding the huge increases in special contributions which have been made, there is no evidence that the situation has been improved by all of this activity. Many of the actions, such as closure to new members, have raised rather than lowered the cost of pension provision. The wide-spread use of derivatives has failed to consider their impact on cash-flow management. In all of the analysis, the most important dimension for the long-term, the income generation of assets, has been overlooked.
In spite of all of these actions and incurred costs, the assets and liabilities of pension schemes do not now co-vary more closely than in previous times; the techniques have been ineffective.
With pension scheme hedging of risk exposures, the position of the sponsor employer who is exposed to all of these risk factors in different ways is not considered; this is highly inefficient. The hedging strategies treat the risks within a scheme as if they are unconditional risks when they are actually predicated on sponsor insolvency. In fact, the sole risk faced by a DB pension scheme is the insolvency of its employer sponsor.
8) Regulatory regimes such as Solvency II should be applied to pension schemes.
This is simply nonsensical. If pension schemes were to be required to hold assets in excess of the technical best estimate of their liabilities to the extent that insurance companies must hold capital in excess of their estimated liabilities, the total bill would be between £330 and £550 billion. This is some three to five year’s total UK business investment. It is some 20% – 40% of the capital of the UK private sector. It is also more than one hundred times the expected annual loss of pension schemes due to sponsor insolvency.
This is a classic situation where insurance of the insolvency risk is the efficient solution. The focus of pension regulation on assets and scheme funding is misconceived. The risk to pension schemes is the product of the likelihood of sponsor insolvency and the degree of underfunding at the point in time when insolvency occurs. The level of scheme funding is immaterial until that insolvency event occurs. It is highly inefficient to focus upon the secondary aspect alone, though it is inappropriate for the regulatory and supervisory authorities to intervene in questions of sponsor viability.
DC with all of its dependencies on financial markets is highly inefficient relative to collective DB. To deliver the same pension outcome it will cost at least 50% more in contribution costs than collective DB. As DC is no more than a tax-advantaged savings scheme, the ramifications of this for taxation policy and social equity are obvious.
The report makes a number of proposals:
Most notable among these proposals is an option for the introduction of unfunded, insured, occupational schemes.
As a book-entry, unfunded arrangement, there are obvious incentives for an employer sponsor to offer pensions. The affordability of pensions rests directly upon their performance. It aligns the interests of employee and employer.
The cost competitiveness of the UK corporate sector and its capital structure would be radically improved.
Employee security is assured by private sector pension indemnity assurance covering sponsor insolvency risk. The cost is a small fraction of costs currently being incurred.
In fact, legacy partially funded pension schemes may also be improved and their costs dramatically reduced by the introduction of pension indemnity assurance covering them. The Exchequer would gain materially from the decrease in tax concessions made for pension contributions.
Lest anyone think this is an academic solution, the fact is that pension indemnity assurance, even for unfunded schemes, exists and has been a significant part of the Swedish pension system for over fifty years.
The full paper is freely available from: www.futureof pensions.org