The Trend Away From Defined Benefit Pension Provision In The UK
by Dr. Ros Altmann
(All material on this page is subject to copyright and must not be reproduced without the author’s permission.)
This note is divided into four sections and the following summarises the main points in each.
SECTION 1: Why the trend away from DB is inevitable:
1. Longer periods of retirement
2. Maturing of DB schemes (older workers require higher contributions than the young)
3. Companies unwilling to continue underwriting an open-ended liability
4. Ending of scheme surpluses and of contribution holidays
5. Asymmetry of tax treatment of surpluses
6. Continuing legislative changes
7. Earnings cap
8. Increasing complexity of UK pensions regime
9. Increased complexity a particular problem for smaller schemes
10. Workers have not really appreciated the value of their DB pensions
11. Employers have to bear all investment and inflation risk
12. Falling equity markets
13. FRS17 – increased volatility, no smoothing
14. EU discrimination legislation by 2006
15. It was perhaps always inevitable that DB would only last while schemes were young
SECTION 2: Why a final salary DB scheme is NOT always a good scheme to have:
1. A DB pension scheme is only as good as the employer who provides it
2. Employers can decide to stop a final salary scheme at short notice
3. DB is old fashioned – not suited to gradual retirement, due to final salary emphasis
4. DB schemes can be seen as a soft target to raid in good times
5. DB pension schemes can put the solvency of the employer at risk
6. DB is not as flexible as DC~
7. DB may be less good for job changers
8. DB is less good for low earners
SECTION 3: Ideas for slowing the move away from DB, to allow time to get DC working better
1. Require employers to explain honestly to the workforce what is happening
2. Require employers to give reasonable notice before changing from DB to DC
3. Require employers who close DB schemes to pay for individual financial advice
4. Suspend FRS17
5. Simplify pensions legislation
6. Workforce could demand companies contribute the same under new DC arrangements
7. Workforce could demand pay increases to make up for reduced pension contributions
8. Change from final salary to average salary, or 60ths to 80ths
9. Incentivise employers to provide a DB scheme
10. Reinstate ACT relief
SECTION 4: What do we need to do to make DC work better?
1. Ensure contributions are sufficient
2. Simplify DC regulations – introduce just one regime for DC pensions
3. Offer better investment options
4. Offer guaranteed products
5. Make sure people get the right annuity and best rates
6. Introduce ‘best practice’ guidelines and ‘safe harbour’ regulations for DC scheme trustees
7. Improve financial education
8. Incentivise employers to give financial education or financial planning advice
At the moment, the UK is by far the best pensioned country in Europe and much of this is due to good final salary schemes. We have built up a strong retirement savings culture, but the situation in early 2002 is putting this at risk. The pensions environment has become so complicated now (after years of continued tinkering, changing regulations, additional burdens on employers, complicated layers of ‘simplification’ bolted on to existing rules) that almost no-one actually understands pensions any more. We have volumes of rules and regulations, we have so many different limits for both contributions and benefits, we have at least 4 different regulatory regimes for money purchase pensions and we are at the point now that, unless something changes, people are going to be put off putting money into a pension at all! This would mean a risk of increasing poverty in old age for future generations and huge rises in government spending to support older members of society who no longer had their own private provision.
I think the start of the 21st Century will be seen as a watershed for pension provision in the UK. The situation must change, to restore confidence in pensions, make DC schemes work better, simplify the whole pension regime and help people save more for their own future. The move away from final salary schemes does not have to be a disaster at all. I believe the demise of DB was always inevitable for private companies and final salary schemes are not always as good as they are currently thought to be. Almost everyone, at the moment, believes that final salary schemes are preferable to money purchase schemes. But this need not be the case. In fact, some final salary schemes are not very good at all. It is possible to argue that the modern world is not really suited to DB anyway, for a variety of reasons.
If we try to force a company which cannot afford it, to keep its DB scheme for too long, this might actually force the firm into bankruptcy and then the members will not only have lower pensions, but they will have lost their jobs too!
The trend away from Defined Benefit has many causes. It cannot be blamed on any one or two factors, it is a combination of circumstances which means that the trend would have been inevitable anyway. The exact timing has been influenced by some particular factors, but it is an unstoppable movement. At the moment, the debate is focussed only on the benefits of DB, but DB does have its drawbacks too. The real problem is not so much that DB schemes are disappearing, but more that DC is not working properly. The crucial issue is that the move to Defined Contribution is handled well.
This note discusses the current trend away from DB pensions and tries to highlight some issues that may have been lost in the recent debate. It is divided into the following sections:
– Why the trend away from DB is inevitable
– The disadvantages of DB
– What might be done to slow the move away from DB?
– How to make DC work better
SECTION 1: Why the trend away from DB is inevitable.
It is my view that final salary pension schemes were never likely to last among private sector employers. I think it was always inevitable that companies would move away from them, but several particular factors have occurred at the same time at the start of the 21st Century, which has resulted in so many employers moving away from DB at the same time. The trend has, however, actually been building up over many years and is due to the following influences:
1. Longer periods of retirement:
The trend to earlier retirement and increased longevity have placed huge strains on DB schemes, in terms of unexpectedly large cost burdens. It was useful, during the 1980’s and 1990’s, for firms to use their pension fund – which was typically in surplus – to fund the costs of obtaining a leaner workforce. Unfortunately, though, the schemes have matured since then, surpluses have whittled away, people are living longer than expected and the costs are coming home to roost.
2. Maturing of DB schemes:
Schemes have grown so large that they are often dwarfing the size of their sponsoring company as, over time, they have become more mature. There are increasing numbers of members now drawing pensions and a smaller relative proportion of younger employees to continue contributing. The older the worker is, the higher the contributions for him or her need to be. As the proportion of older workers and pensioners increases, this again places much larger burdens on schemes than was the case in the past.
3. Unwillingness to continue underwriting an open-ended liability – no certainty of cost:
Running a DB scheme entails not only high costs, but increasing uncertainty of those costs. The recent trends in corporate financial management are focussing on gaining more certainty of cost, better planning and budgeting and better overall cost controls. This is very difficult to do with DB. Employers cannot reliably predict any of the factors which influence how much the running of their scheme will cost. They do not know what investment returns will be, what salary inflation will be, what longevity will be, what government regulations will be, etc.
4. Ending of scheme surpluses and contribution holidays:
Until recent years, DB pension schemes were often in surplus and employers were enjoying ‘contribution holidays’, so that the schemes were not really costing them much anyway. This situation is no longer very common and perhaps it was always inevitable that there would be a trend away from DB when the schemes became more mature, when investment returns were poor and the companies started to have to make bigger contributions.
5. Asymmetry of tax treatment of surpluses:
To some extent there has been a problem for DB schemes in that they are not treated equally on surpluses or deficits. If the scheme is over funded (i.e. in surplus) by more than around 5%, the surplus over this level is taxed. However the scheme has to make up any and all shortfalls in full. So the scheme cannot keep all of its surplus in the good times, but has to make up 100% of the deficit if things go wrong. This could prove a deterrent to employers, could prevent optimal investment decisions, could mean that companies are more willing than they otherwise would be to pay out more of the surplus than they should and not to keep back enough to cover the more difficult times.
6. Continuing legislative changes:
Over the years, there have been so many legislative changes imposed on DB pension schemes, that the costs of compliance and of actually paying pensions has risen enormously. Many of these changes were actually designed to protect members of these schemes, but, unfortunately, there have been so many of them that they may well have contributed to radically reducing members’ security, by taking away the employer’s guarantee. The changes were introduced in an environment over the last 20 years in which pension schemes had surpluses and could afford to improve the pensions offered. The changes have included:
escalation requirements – such as LPI
MFR and associated compliance costs
member nominated trustees
pension sharing on divorce
SRI – Socially Responsible Investment
SIP – Statement of Investment Principles
7. Earnings Cap:
Some have argued that the fact that top executives can no longer put unlimited amounts into the pension scheme has caused several problems for DB schemes. Firstly, it may prevent labour mobility. Secondly, it may mean that the Directors of a company are less committed to running a DB scheme, because they cannot benefit from it as much as before. I am not convinced by this argument, especially since the earnings cap was introduced in 1988, but the trend away from DB started much later and, in any event, Directors are able to keep a DB scheme for themselves and close it to new members!
8. Increasing complexity of pensions regime:
DB pension schemes have become hugely complex to run. There are compliance requirements, different tax regime requirements, different escalation requirements, etc, depending on what period the member’s contributions relate to. There have been so many pieces of major and minor pension legislation which schemes have had to cope with, that it has become hugely expensive to run each scheme.
9. Increased complexity has been an enormous problem for smaller schemes:
The high costs of compliance, administration, legislative changes and investment advice or management are very onerous for smaller schemes, and do partly explain why so many smaller schemes have wound up in recent years.
10. Workers have not really appreciated the value of their DB pensions:
Until recently at least, workers tended not to focus on their pension entitlements much. The fact that employers were paying large amounts into a company scheme on their behalf was not really appreciated. Effectively, these contributions and pension promises are a form of deferred pay, but the costs are high and not really understood by the employees.
11. Employers have to bear all investment risk and inflation risk:
With DB, the level of benefits paid still needs to be maintained, even if investment returns are low and regardless of inflation effects. These are big risks for employers to bear.
12. Falling equity markets:
The falls in equity markets over the last couple of years have hastened the demise of DB schemes, because of the implications for funding, surpluses and investment returns.
The recent change in accounting standards, which requires changes in pension scheme surpluses or deficits to be shown on the balance sheet, has coincided with sharp falls in equities. With FRS17, there is no longer an ability to smooth market movements over time and all liabilities are valued in relation to AA corporate bonds, which has inflated liability values relative to past levels. Thus, the asset base has fallen and the assumed liabilities have increased, so pension funding appears much worse than before. This issue can no longer be smoothed or hidden away and is being put on the boardroom map. It is another reason for employers to be frightened of maintaining DB schemes. Of course, if equity markets were to recover sharply, the funding position would improve and FRS17 would be less of a problem, but the timing has been quite unfortunate.
14. EU discrimination legislation by 2006:
The UK will need to comply with the EU Directive which prevents any discrimination on grounds of sex and age. This could have a significant impact on DB schemes because contributions are different for different age groups and pensions are lower for women than men. The costs of complying with this Directive would probably spell the end of DB by 2006, but their demise has been happening sooner for all the other reasons listed here.
15. Was it always inevitable that DB would only last while schemes were young anyway?
It could be argued that DB schemes among the private sector in particular, were never going to last and that it was a little naive of people to expect that employers would continue to underwrite these kinds of open ended liabilities once the schemes grew large. It is easy and relatively cheap to run a DB scheme when your workforce is young, there are not many pensions to pay and investment returns are good. Contributions and costs rise sharply as the proportion of older members or pensioners increases, so perhaps the advantages of DB are only temporary and cannot be sustained in the longer term. Not only this, but the government has given big tax relief to the better off (i.e. the most important members of the workforce) to encourage pension provision. The tax relief is up front and is only recouped by the Government on retirement (if at all). Of course, once the number of pensioners increases and the proportion of workers diminishes, the costs of running a DB scheme become much more obvious and onerous. The precise timing of when the burdens become too big is impossible to predict, but it could be argued that it would happen at some point anyway. Even if one does not accept this argument, it certainly seems to be the case that the maturity of schemes, coupled with all the other burdens that have been placed on DB, have made it very difficult to justify continuing this kind of paternalistic provision.
SECTION 2: The Disadvantages of DB
DB does have disadvantages and it is not always the best type of scheme to have. All the comment about pensions at the moment implies that a final salary scheme is the best type of scheme and that DC schemes are inevitably inferior. This line of argument is not necessarily accurate. Some DB schemes are not worth having and DC can actually work very well – although the way it is operating in the UK at the moment is certainly not optimal.
1. A DB pension scheme is only as good as the employer who provides it:
A DB pension from a weak employer may end up being worthless. At the moment, even with the MFR and all the safeguards, members of a DB scheme whose employer fails, may not even get their own contributions back! The law protects pensions in payment in full and deferred pensioners also come before active members. If there is not enough left in the pension fund after these payments are guaranteed, members could get nothing. There is, therefore, no guarantee of security for the current workforce in a DB scheme.
2. Employers can decide to stop a scheme at short notice:
Workers who are in a DB scheme could suddenly find their employer has wound up the scheme or become insolvent and, if they are older, they will have little time to make up extra contributions. They may not fully realise what is happening, and they will have lost valuable life insurance and disability benefits which might be hard to replace later in life. They may, therefore, actually have been better off if they had been in a DC scheme earlier.
3. DB is old fashioned – not suited to gradual retirement, due to final salary emphasis
The old-fashioned, paternalistic idea of the employer supporting his or her workers throughout their retirement is quite outdated. The notion of relating pension to final salary is also not ideal for today’s workforce. It assumes that people will retire when they reach their peak earnings and have progressed to the top of their potential. But, as people are living longer and health status has improved, there are many advantages in the idea of ‘gradual retirement’. Instead of the traditional idea of working full time until a specific age and then suddenly stopping altogether and becoming ‘retired’, it makes more sense to consider gradually cutting down on working time at the end of one’s career. Perhaps going down to 4 days a week, then 3 days, or mornings only, or job-sharing for older workers. Employers would have the benefit of loyal workers who have built up firm-specific skills and who could train or mentor younger workers. Employees would be able to gradually increase their leisure time, rather than suddenly stopping work altogether. Most people would not want to continue working full time, but would be happy to consider part time opportunities (or even some re-training or job re-definition) to be able to earn an income for longer, keep in contact with their colleagues and feel useful, rather than suddenly ‘too old’ to contribute to society. But this would mean that a person’s ‘final salary’ would not be their peak salary and the final salary pension scheme is not suited to this kind of potential modern flexibility in the labour market. The money purchase concept is much more suited to this.
4. DB schemes can be seen as a soft target to raid in the good times:
Governments are often tempted to impose extra burdens in the form of taxes or regulations on DB schemes when they are in surplus. In addition, employers have been tempted to use the pension fund as a soft option for engineering a reduction in the labour force without incurring too high a cost. Such actions weaken the scheme and endanger its long term viability. However, in the meantime, workers are lulled into a false sense of security that they are in a good scheme. Of course, surpluses are really there to provide a cushion in the bad times (such as the last couple of years) but, if they are whittled away when they build up, they will not be there to provide the necessary backing when they are needed.
5. DB pension schemes can put the solvency of the employer at risk:
If a DB pension scheme runs into funding problems, is quite mature and is backed by a weak employer, the costs of maintaining the scheme can ultimately force the sponsoring company into insolvency. This would be double jeopardy for the workforce. Firstly, they lose their job and secondly their pension is much reduced.
6. DB is not as flexible as DC:
Workers cannot choose a retirement age or obtain 10 year capital protection in a DB scheme.
7. DB may be less good for job changers:
Even after the legislative changes designed to protect leavers, transfer value calculations are based on methodologies that are far from transparent, so people will often lose out when they change jobs.
8. DB is less good for low earners:
Top management and those who progress up the salary ladder fast will get big pension payments on a final salary DB arrangement, but those on more modest incomes, who do not progress through the ranks, will be subsidising these payments to a certain extent. People on shift work, with overtime payments, commission and bonuses may not have the full value of their salary payments reflected in their pension in many cases. Contributions are a percentage of pay and are therefore best for the rich.
SECTION 3: What might be done to slow the trend away from DB?
There are various measures that could be proposed, to try to slow the trend away from DB. These measures are unlikely to reverse it, but would at least allow more time to get DC working better and time for workers to plan better for their future under DC. The trend is happening rather fast and there are advantages in trying to slow it down, but I do not believe it can be stopped altogether. These are suggestions for consideration by policy makers, if they want to address some of the immediate concerns about the pensions situation.
1. Require employers to explain honestly to the workforce what is happening:
Government could require employers to explain to their workforce exactly what is happening to contributions and non-pension benefits, when they move from DB to DC. At the moment, many are being disingenuous in the way they are describing the move. Even suggesting it is a good thing for everyone. If employers are effectively cutting contributions in a new DC scheme, this should be explained to people, partly for the sake of honesty and, more importantly, so that people have the option of contributing more for themselves. If they do not realise that contributions are so much lower, they will not be alerted to the need to make more provision themselves. In addition, the DC schemes often offer lower life assurance and disability insurance benefits, and this should be explained clearly to the workforce.
2. Require employers to give a reasonable period of notice – perhaps 12 months – before changing from DB to DC:
Currently, not much notice is required, but, if people had time to plan for the change, at least they would have a chance to make the necessary arrangements for their future security. Requiring employers to give a year’s notice would allow some time for adjustment.
3. Require employers who close their DB schemes to pay for individual financial advice for members:
At the moment, members who are switched into DC schemes, or new members joining have no idea how much they need to contribute or how to plan to get a decent pension. If employers are cutting costs by moving to DC, perhaps it is not unreasonable to suggest that they offer their workforce some time with an independent financial adviser, who could help them work out what contributions they need to make, to achieve a desired level of pension when they retire.
4. Suspend FRS17:
In the short term this is probably one of the most powerful ways of slowing the trend away from DB. The rationale for suspension would be that there is likely to be a new European pension accounting initiative introduced around 2005, which may be different from FRS17. It would make sense to wait and see what the European requirements will be, so that UK companies do not have to change now to FRS17 and then again to some other standard a couple of years later.
5. Simplify pension legislation:
The opportunities for simplification are enormous and proper reduction of administrative and legislative burdens would be of great help. The possibilities here are too numerous to list!
6. Workforce could demand companies should contribute the same amount into DC as DB:
It may be possible for workers in some companies to demand that employers, if they introduce a new DC scheme, should contribute at least as much as they had done into the DB scheme.
7. Workforce can demand an increase in pay to compensate for loss of DB:
The move to DB, if accompanied by a reduction in employer contributions, is effectively a pay cut for the workforce. If unions and workers generally realise this, and start to demand an increase in pay to make up for it, employers may think twice about making the switch. At the moment, it seems like a one way option for them. They can gain certainty over their costs and cut costs at the same time as reducing the burdens of running the DB scheme.
8. Allow average, rather than final salary, allow 80th’s rather than 60th’s:
Some technical changes to reduce the cost of providing DB pensions have been proposed. These may be of interest to some companies, but would entail a reduction in value of pensions promised. It is argued that a smaller DB pension is still better than moving to DC. This is a debateable point.
9. Incentivise employers to provide DB:
The Government could offer bigger tax breaks to companies, if they provide a decent DB scheme, or it could give tax relief to Directors of companies who offer a generous DB scheme. This proposal could help to keep some DB schemes, but it does risk distortions. Directors may only offer good schemes to get their own large contributions and tax relief, even if the company cannot afford them, but the Directors may have retired or moved on with big transfer values before the problems surface!
10. Reinstate ACT relief:
The removal of ACT relief certainly added an extra burden on pension schemes. It was affordable in times of pension surpluses, but has become more difficult to justify recently. However, on its own, reinstating this relief would probably not be enough to change the mind of a corporate board looking at switching to DC.
SECTION 4: Some ideas to make DC work better
The trend to DC need not be as bad as is currently feared. Of course, if employers are just using it as an excuse to cut contributions and the workforce does not demand any compensation, there is a net loss to the employees. At the moment, DC pensions are normally inferior to DB from a good company, but it is not difficult to recommend ways of improving DC and making it work well.
It is vital to do this for many reasons. Importantly, although people portray the move from DB to DC as transferring risk from the employer to the employee, ultimately the risk is actually transferred to the State. If people do not get enough from their DC pensions, more of them will need to be supported by the taxpayer. The current perception is that the UK is the best pensioned country in Europe. This has caused a certain amount of complacency among policy makers about the need for action on pensions. But we have probably reached a watershed period in the UK. Estimates of public expenditure on pensions have suggested that State costs of supporting the elderly are well under control. However, these forecasts have not factored in the risk that DC pensions will not be as generous as DB and will cause more people to need State support. The introduction of the MIG and Pension Credit, coupled with the switch away from DB, will probably mean State spending on support for the elderly will rise way beyond current forecasts.
We obviously need to remove the current disincentives to saving that exist in the UK, in order to get more people providing for their own future. It would also be helpful if we could introduce better incentives to save for those who currently do not do so. The system of tax relief offers much more incentive to the rich, who are most likely to save anyway. Middle Britain probably needs more encouragement and reward for making the income sacrifices now, that will provide for their security in old age.
These are the issues that need to be addressed, and would help DC provide decent pensions:
1. Ensure contributions are sufficient:
A most important issue is to make sure that contributions are high enough to provide a decent level of pension. Removing saving disincentives and improving incentives for middle income groups would be of great help here. If everyone was offered the same tax incentive, instead of more for the rich, more people would be incentivised to save. Perhaps giving everyone 50% tax relief on the first, say, £1500 a year that they pay into their pension and then everyone getting 40% relief on the next £1500 and so on, with a sliding scale of tax reliefs on larger amounts, the same for everyone. This would be much fairer than the current system. But it is also necessary to make sure people are better informed about how much they actually need to save to provide their desired level of income in retirement.
2. Simplification of DC regulations – introduce just one regime for DC pensions:
At the moment, there are at least four different regulatory regimes for DC pension arrangements. The environment has become so confusing that it is causing unnecessary complications for people who want to save for their retirement. Different contribution limits, different rules on how you are allowed to take the retirement benefits, partial concurrency, different investment restrictions, complex stakeholder ‘non-relevance’ tests and so on, are all compounding to make DC pensions a dangerous minefield for most people. We really need to merge all the old regimes and come up with one single DC regime. This could perhaps take the best features of all the existing regimes and make it clear what contributions are allowed, what investments are permitted, how the pension can be taken, with common standards for all types of DC pension. At the moment, the rules are different for AVC’s, FSAVC’s, personal pensions pre-1988, post- 1988, stakeholder, GPP’s, SIPP’s, SSAS’s etc.
3. Offer better investment options:
Currently, investment options in DC schemes are not working as well as they could. Many occupational schemes offer no choice of investment – just one option. Some only have active and not passive funds. Some use DB benchmarks, which are not appropriate for DC liabilities. Some offer lifestyling options, which switch people into bonds in the 10 years or so before normal retirement age. But, if a person has no other equity assets, is not planning to retire at normal retirement age or goes into drawdown on retirement and has to buy back the equities sold before they retired, this option will not provide the best pension. Lifestyling is probably more appropriate post retirement, rather than pre-retirement.
4. DC could offer guaranteed products:
Just as there are products in the market place which currently offer guarantees of certain levels of return, or at least protection of capital, such products would probably be very useful in terms of pension planning for DC pensions. This type of product, for at least part of a person’s pension savings, would be helpful, in order to achieve some level of certainty over the amount of pension one could achieve from DC.
5. Make sure people get the right annuity and the best annuity rates:
Improving the workings of the annuity market will help DC provide better pensions.
6. Introduce ‘best practice’ guidelines and ‘safe harbour’ regulations for DC scheme trustees:
In a DB scheme, the investment performance will affect the security of the pension. In a DC scheme, the investment performance will affect the amount of the pension. It is, therefore, arguably more important for trustees of a DC scheme to be able to make good investment decisions and offer good investment options, than in a DB scheme. It is not clear that trustees have focussed clearly on this responsibility and it would be helpful if the authorities issued some ‘best practice’ guidelines and, possibly some ‘safe harbour’ regulations, to guide trustees and improve the investment profile of DC schemes.
7. Improve financial education:
It is essential that people are better informed about how much they need to save to provide their desired level of income in retirement. The FSA is making great advances in providing literature and helping people understand financial matters. More still needs to be done.
8. Incentivise employers to provide financial education and financial planning advice:
It would be helpful if employers could offer workplace education courses, to help people understand what is required of them. It would also be helpful if employers could even provide proper independent financial advice for their employees. At the moment, the cost of this is not tax deductible and the employee may be taxed on the advice as a benefit in kind. Permitting employers and the self employed to offset the cost of advice against their tax liabilities would make it far more likely that it would be taken up. People would then be more likely to find out how to and how much to save for their future. (Exempting financial advice from VAT would be of further help!)
In summary, the current movement away from DB pension provision is probably unstoppable. It does not, however, have to mean a disaster for pensions, if the change to DC is handled well. In fact, there can also be problems with having a DB pension, since such schemes are only good if the sponsoring company is strong.
There can also be benefits of having a DC pension, as long as contributions are sufficient, investment returns are satisfactory, annuity markets work better and regulatory issues are addressed. DC can be made to work much better than it currently does. It would be useful if the current disincentives to saving are removed and better incentives are introduced, in order to ensure more people provide for their future. This should be more the focus of the current debate on pensions, rather than trying to reverse the switch away from DB. Let’s make DC provide better pensions for everyone.