|
Bookmark this page www.oecd.org/asktheeconomists
See below the questions and answers from this online debate that took place on Thursday 5 April with OECD economists Monika Queisser, Ed Whitehouse, Fiona Stewart and Pablo Antolin.
Q. The UK has devised the Personal Accounts system to cover those who have no pension provision. Is this the best way forward for the UK to tackle its pension crises?
Raji Menon, UK
A. As income levels guaranteed by public pensions decline, many governments are encouraging individuals to supplement this income with private pension savings. However, it has proven difficult to raise the coverage rate of private sector pensions above around 50% of the workforce without some form of compulsion. This is the challenge which the UK is also facing.
Individual accounts are an innovative way of increasing pension coverage, and have been used successfully in Latin American and Eastern European countries. However, these systems are mandatory and it will therefore be interesting to see whether the 'automatic enrolment' approach (where by individuals join the system by default and have a period in which to 'opt out') - which is being tried in the UK and in New Zealand - will have similar success.
It should, however, be noted that individual accounts do not solve all the problems of pension coverage, as they are only applicable to those members of society who are in the workforce. Many countries, such as Chile, still find that a large percentage of the population without work, the self-employed or informal workers do not have any pension provisioning. Hence proposed reforms to the individual account system in Chile include making accounts compulsory for the self-employed and providing subsidies for young workers with low incomes to encourage participating in formal employment.
Q. Some people believe that notional defined contribution (NDC) pensions would solve the problems of pensions. Which countries have NDC pensions? What does the OECD think of them?
Montserrat Pallares-Miralles, World Bank
A. Italy, Latvia, Poland and Sweden have adopted notional accounts, or NDC pensions.
Notional accounts are a kind of publicly provided pension where the benefit depends in individual earnings. Individuals' contributions are credited to a notional account, which is notional in the sense that the money does not remain there. Each year, notional interest is added to the account, usually depending on some macreconomic variable, such as the rate of growth in the economy or growth of average earnings. At retirement, the balance of the account is transformed into a periodic pension payment using actuarial calculations that depend on life expectancy.
Notional accounts have a number of desirable characteristics. First, the pension benefit depends on lifetime earnings and not on a shorter period of final salary. It is more equitable and distorts the labour market less to base pensions on lifetime earnings. Secondly, the pension benefit depends on the age at retirement, with smaller pensions for earlier retirement and larger pensions for retiring later. Thridly, the pension benefit is linked to life expectancy, so as life expectancy increases, pensions are redcued or people have to work longer to get the same pension.
However, notional accounts are not necessary to achieve these goals: any well-designed public pension scheme bases pensions on lifetime salaryb and has adjustments for early and late retirement. And many countries have introduced links to life expectancy in traditional kinds of public pension scheme: including Finland, Germany and Portugal. Also, notional accounts are not sufficient to achieve these goals: it depends on their precise design.
Q. Which country has the best pension system?
Andrew Clark, UK
A. This is a tricky question. There are many elements of OECD countries' pension systems that work well and can serve as "best practice". But there is no single model that can or should be applied in every country. OECD pension systems developed gradually and were strongly influenced by each country's economy, societal values and cultural norms. What works well in Switzerland or Sweden will not necessarily be the best solution for Mexico or Japan.
It also depends on the criteria one uses to assess the best system. Is it the system that offers the highest pension replacement rates? In this respect, the earnings-related pension systems in Greece and Luxembourg would be the best but they would score much lower on financial sustainability. Or is it the system that is most targeted towards the poor? With this criteria, the universal flat-rate pension schemes of Ireland and New Zealand or the basic pension in the UK would score highest. Another criteria could be diversification of retirement income source. On this count, Sweden would score the highest since pensioners there receive benefits from 5 different pension system components, funded and pay-as-you-go, defined-benefit and defined-contribution, public and private. Having the best of all worlds will require determining the various objectives of the pension system and conducting a careful analysis of the different countries' experiences. The OECD tries to contribute to this through its work on public and private pensions.
Q. Do individuals understand the need to save for retirement and how best to do that?
Ingeborg Scheven, Germany
A. Surveys across OECD countries and worldwide consistently show alarmingly low levels of financial literacy in general and understanding of the need and importance of saving for retirement in particular. For example, surveys in the USA have shown that four out of ten workers are not putting any money aside for retirement, whilst a report in New Zealand found that many individuals are either 'unwilling or not able' to save enough for retirement, with around 30% of households spending more than they earn. 71% of respondents in a Japanese survey had no knowledge about investment in equities and bonds and in Canada, respondents considered choosing the right investments for a retirement savings plan more stressful than going to the dentist!
Though measuring the impact of financial education is challenging, programmes can play a role in improving individuals' understanding of the changing pension environment, of the need to save and how to invest. For example, in Germany providing information on pension entitlements has been found to play a significant role in additional planning for retirement. In the USA participation in 401(k) schemes rose 12% after financial education was provided to workers and contribution rates also increase, whilst a study in Chile estimates that self-employed workers with an above average knowledge of the pension system are 14% more likely to make contributions.
OECD governments are consequently increasing realising the importance of financial education, particularly in relation to retirement savings, with national campaigns having been launched in the UK, Australia, New Zealand and the Netherlands, amongst others. The OECD will continue to provide international experience and analysis on this important subject.
Q. In my opinion the shift from income to consumption taxation we observe in many European countries is not only motivated by efficiency issues (which are not that clear and convincing in economic theory).
Don't many governments use it to shift the tax burden by stealth more to pensioners? Currently income taxation in most countries is quite modest on pension incomes and as it is politically very hazardous to decrease pensions, shifting to consumption taxation is a way to make pensions less genereous by stealth.
Is this a plausible interpretation and do you as economists for the OECD have a clear view on the efficiency and equity issues involved by this shift and its intergenerational impact (in economic theory the Ramsey analysis, the Corlett and Hague Theorem and Atkinson-Stiglitz Separability Theorem are seminal but don't give a lot of practical help).
Patrick Broullard, France
A. There has been a global shift to consumption taxes, not only in Europe. For example, Australia, Canada, Japan and Switzerland have adopted national general consumption taxes (like value-added tax, VAT) as have many developing countries. The shift from direct to indirect taxes has many explanations. One is the efficiency argument: taxing consumption reduces the need for direct taxes on labour and capital.
The effects of this shift are complex and it is important to take a lifecycle perspective. For example, income taxes tend to tax some or all of savings, which consumption taxes do not. But savings are a means to future consumption, and so consumption taxes will be paid in the future when the monmey is spent. So a shift to conumption taxes may increase the tax burden on pensioners, but when the change is fully in place, they will have paid less in income taxes when they were working.
Many countries do, indeed, offer tax concessions to older people meaning that the effective burden of income tax on pensions is small. However, it is difficult to justify the fact that a pensioner on the same income as a worker pays less in tax. Most income tax systems are progressive, so pensioners would pay less in income tax anyway even without special concessions.
Cuts in pension benefits for future retirees in recent pension reforms have, in many cases, been large: in the 16 OECD coutnries that have had major pension reforms since 1990, benefits for a worker entering the labour market today are, on average, 25% lower than they would have been without reform. This dwarfs the effect on pensioners of the shift from income to consumption taxes.
Q. Several OECD countries have recently raised their retirement ages in the public and/or private sectors. Are there examples of good practice in OECD countries for encouraging workers to keep working longer? And if there are, are there any studies that show a truly positive impact of these measures?
Anne Gosselin, Canada
A. Increasing the effective age of retirement, which has fallen despite rapid increases in life expectancy and healthy life expectancy, would be a great help is dealing with the demographic burden of population ageing. Such increases have a double positive effect, potentially both reducing benefit expenditure and increasing tax and contribution revenues.
The OECD's analysis has shown that many countries continue to subsidise early retirement, through pension systems and other pathways to early retirement. In old-age pension systems, for example: (i) some allow people to retire early with little or no reduction in benefits to reflect the longer duration for which pensions will be paid; (ii) resource tests encourage low earners to retire as early as possible; (iii) some systems give little or no extra benefit for an extra year's contributions; (iv) earnings tests prevent peopel combining work and pension receipt. The OECD's approach can therefore equally be summarised as stopping subsidies for early retirement as much as coercing people to work longer.
There is widespread evidence that pension incentives affect people's retirement decisions (along with many other factors, such as health, the state of the labour market etc.) Recent reforms to improve retirement incentives have had a clear impact on employment rates of older workers in Finland, France, New Zealand, the United Kingdom and the United States, for example.
Q. Across Europe, a growing number of pension funds are being driven to adopt more conservative strategies by regulation and changes to accounting standards. Isn’t this defeating the purpose of maximising returns to the ultimate beneficiary that is the pensioner?
Raji Menon, UK
A. The 'perfect storm' in the pension industry at the start of the Millennium saw declines in equity markets caused pension funds’ assets to decline sharply. At the same time the promises made by the pension funds, their liabilities, rose as interest rates fell – pensions, being long-term ‘debt-like’ promises, become more expensive when interest rates are low. This underfunding, which still haunts many pension funds, provoked reactions from governments and regulatory authorities in the form of stricter funding and accounting rules, requiring pension funds to match their assets and labialise more closely and plan sponsors to declare pension deficits on their balance sheets. Such regulatory changes have been effecting pension fund balance sheets - for example with allocations to equities declining and increased demand for long-term, index linked government bonds.
Pension funds always have to strike a delicate balance between managing risks and maximizing potential higher portfolio returns. The new regulatory standards do indeed tip the balance more in favour of risk reduction, but the pension funds reaction to these regulatory changes have been interesting. In addition to increasingly matching their assets and liabilities, pension funds are also exploring ways to manage their portfolios more efficiently - through a focus on absolute return investing, increasing diversification (e.g. investment in real estate and commodities) and through exploring new, alternative investments. It is hoped that the combination of these regulatory, risk management and investment developments can deliver greater retirement income and security to pension beneficiaries in the future.
Q. Is there any likelihood of mortality-linked Government Debt being issued in the foreseeable future? What impact might this have on reducing pension scheme risk? Perhaps a multi-national life insurer may be able to offer such debt?
Paul Bowden, UK
A. Governments could support and encourage the development of a market for mortality or longevity-indexed bonds where pension funds and participants in general would hedge against longevity risk. Governments could achieve this by issuing longevity indexed bonds and by producing a longevity index. Unfortunately, this public policy role is hampered by the fact that governments are themselves already exposed to significant longevity risk through their public pension systems. Such instruments are therefore unlikely to be issued by governments in the near term. However, governments could take other steps to help develop a market for these bonds, such as producing a longevity index. In terms of private sector issuance of longevity bonds, EIB-BNParibas did attempt to launch such an instrument last year in the UK market, but it did not succeed in finding enough demand (partly on account of the structural issues such as the bond being limited to 25 years and other more pressing regulations to attend to).
Longevity index-bonds would have an impact on reducing pension scheme risk. For example, an individual aged 65, using current life expectancy (rising 1 year per decade), with interest rates at 5.5% would expect to receive an annuity payout of 100. If his life-expectancy were to rise by 2 years per decade, the annuity payout would rise to 102.4. Longevity-bonds could therefore help pension providers to hedge against the risk of an extra 2.4% payout. However, the impact of longevity risk should not be overstated. For example, the same individual could expect an annuity payout of 118.6 if interest rates declined to 3.5%, even if longevity does not change, showing that hedging interest rate risk is of more importance to pension providers. What is interesting though is how interest rate and longevity risk interact with each other. If interest rates were to fall to 3.5% and longevity were to increase 2 years, the individual’s annuity pay out would be 122.3. What is also important to note is the magnitude of the longevity effect increases substantially for younger individuals.
See data here
Q. The OECD in March 2007 released a report on reforming public employee pension plans. One of the report's key findings was that many OECD countries face significant costs related to public employee pensions. Specifically, OECD reported that member countries spend on average nearly 2% of GDP on public sector pensions.
While the report identifies several common strategies for reforming public employee pension plans, it seems one important issue is the extent to which enhanced financial information affects policy choices. To what extent, do OECD countries estimate the long-term costs of public employee pensions? Do countries develop accrual based estimates of future pension payments? How has enhanced financial information on projected public sector pension costs affected reform proposals in OECD countries?
Michael O'Neill, The United States Government Accountability Office
A. Most OECD countries have different pension provision for public- compared with private-sector workers: only in the former-socialist countries in Central and Eastern Europe, such as Hungary, Poland and the Czech and Slovak Republic, is there a single, national programme. Expenditure on public-sector pensions is, on average, around a quarter of total public spending on pensions.
Most countries do include public-sector pensions in projections for pension spending: see, for example, the recent projections carried out for EU member states.* The use of accrual accounting is not particularly widespread: pension payments are generally financed on a pay-as-you-go basis from current budgets. Therefore, it is difficult to assess the affect of accounting procedures on the process of pension reform.
* European Union, Economic Policy Committee (2006), “The Impact of Ageing on Public Expenditure: projections for the EU-25 Member States on pensions, health care, long-term care, education and unemployment transfers (2004-2050)”, European Economy, Special Reports No. 1/2006.
Q. As public pension systems are being scaled down, people will have to save more on their own. But workers have the choice between saving in assets such as real estate or a stock of financial assets or buying often tax-favoured retirement products that mandate annuitisation. What are your views on these choices?
Thomas Durand, France
A. People can make financial arrangements for their retirement in different ways. The evidence from most countries shows that workers are often myopic, they almost consistently underestimate their financial needs in retirement and many think they will live much shorter than is the case. In an annuitised financial product, there is no risk that people may outlive their savings: payments are guaranteed until the beneficiary dies. But if there are non-annuitised accounts, there is the risk that a retirees will run out of money. This is why many governments insist that at least part of the tax-subsidised retirement savings be paid out only as an annuity.
The answer to the question also depends on what the public mandatory pension system offers in the respective country. If pensions from the public system are relatively high, people may not need more annuitised income. They may prefer in this case to have more savings that they can use to spend on travel, a secondary residence or other uses. Allowing people to buy voluntary contracts only with annuitisation may result in a situation where pensioners are "over-annuitised". But in systems which offer only a very basic public pension benefit, policy-makers would probably want to limit the risk that pensioners spend all their money and fall back on safety-nets.
Finally, the choice is probably also a personal one. In France, people seem to prefer life insurance to the new PERP product, since tax-subsidised life insurance savings are only bound for 8 years and then become available again while the PERCO must be taken as an annuity. For the moment, this is probably a rational choice since replacement rates for today's retirees are still quite high. But younger workers can expect much lower pensions in the future; many of these workers should think about making safe additional retirement arrangements today.
Q. In the base of defined contribution system, how can we keep compatibility between stability of old lives and financial stability?
Kim, Kun Min from the Ministry of Government Administration and Home Affairs, Korea
A. Given the rising costs of providing pensions (due to increasing longevity and lower returns on assets) many defined benefit pension providers have been forced to restructure their schemes and have chosen to introduce defined contribution (DC) style pensions. Under such schemes only the contributions are fixed, with no particularly benefit being guaranteed - hence they are seen as more 'financially stable' for providers. However, these schemes transfer all risks, such as investment risk and longevity risk, onto individuals - raising the question of whether such schemes can provide a stable pension income in old age.
Yet DC schemes do not have to be inherently risky, and indeed investment and longevity risks can be managed. During the 'accumulation phase', when pension assets are being built up, investment can be made in secure, stable return assets, or individuals can shift the balance of their portfolio from higher risk, higher return assets to more secure investments as they approach retirement age. This so called 'life cycle' approach is the mechanism used in many of the Latin American pension systems, for example. Likewise, annuity products exist to provide a stable income during the 'pay-out phase' when individuals start drawing on their pension income. Policy makers can and should be encouraged to ensure that a regulatory environment exists which allows for stability during both the accumulation and pay-out phases. Financial education and awareness should also be highlighted to help individuals understand and manage risk effectively.
Additionally, further stability can be provided within a DC context by the sharing or pooling of risks. For example, 'hybrid' schemes are developing in some OECD countries, which involve some form of guarantee, allowing risk to be shared by the pension providers and the beneficiary (e.g. the plan sponsor or provider guaranteeing some return on pension assets). Meanwhile in countries such as the Netherlands 'collective DC' schemes exist where the members or the pension fund pool risks between them (by sharing out investment returns). Annuity products provided by insurance companies likewise pool longevity risk in the payout phase.
Q. Among all of the market failure arguments for government intervention in the area of pension provision, the most convincing one in my mind is the dearth of annuities available through private providers. Without mandates, the problems of adverse selection along with preference for flexibility among most consumers are enough to drive this product to extinction. Add the supply side factor of a lack of long term instruments for hedging the risk that life expectancy will continue to increase to unprecedented levels and there seems little hope for the market to satisfy those who value longevity insurance. Thus, the mandates in DC schemes that aim ensure that the public policy objective of consumption smoothing is not compromised at the end of the game when the retirement savings of a lifetime are gambled, spent or lost long before the end of consumption.
Accepting that a mandate makes sense, at least to a minimum level, what is the rationale for private provision? Insurance companies still don’t have appropriate hedging instruments. Prudent companies will still feel obliged to take into account the risk of excess life expectancy that may result from medical advances (as is already becoming an issue in the UK market, for example) and will therefore raise their commissions. Less prudent firms might take on too much risk. The government mandate and the social consequences of default suggest an implicit government guarantee.
There is also much less rationale for allowing much differentiation of products given the straightforward public policy objectives. In contrast, the accumulation period could allow for greater variation in the risk-reward tradeoff within limits. The rationale for competing providers would seem limited to quality of service. Is this enough to justify competing annuity providers in a mandated system?
The alternative would be that governments either took on this function themselves or conducted an auction for a no-frills annuity product with an explicit guarantee covering extreme cases of longevity increase. The economies of scale in administration would reduce costs as would the reduced need for government regulation and supervision of providers.
The question then is whether and in what circumstances would it make sense for governments to play the role of annuity provider, either directly or through outsourcing on behalf of the members of a mandated DC pension scheme?
Robert Palacios, World Bank
A. As discussed in an earlier question regarding longevity bonds, the government is already exposed to longevity risk as it provides annuities in the context of public pension systems. Consequently, it is unlikely that governments will take more longevity risk onto their balance sheets by providing annuities in the private pension context. However, governments could encourage the development of annuity markets by producing longevity indices that could be used by the private sector to issue longevity-indexed bonds.
On the demand side, governments could help stimulate the market for annuity products by improving financial education and thus increasing the understanding of the benefits that these products can provide.
Background reading
Pensions at a glance
Pensions Panorama - Retirement Income systems in 53 countries
Solving the pensions puzzle (policy brief)
Pension challenges and pension reforms in OECD countries
Global Pension Statistics
Ageing society and pensions
Read the questions and answers from recent debates
Made in China. Is the game changing?
Russia: will the boom bust?
Grappling with the world's new IT giants
|
