amsterdam2016.eurofi.net

Ageing population: key challenges posed for the financial sector

Speakers

William R. White
Bank for International Settlements (BIS) , Previously, Economic Adviser and Head of the Monetary and Economic Department
Pablo Antolin-Nicolas
Organisation for Economic Co-operation and Development (OECD) , Principal Economist, Head Private Pensions Unit, Financial Affairs Division
António De Lecea
European Commission , Principal Advisor, Acting Director, Investment, Growth and Structural Reforms
Felix Hufeld
Federal Financial Supervisory Authority, Germany (BaFin) , President
Hanno Mijer
Zurich Insurance Company Ltd. , Chief Executive Officer Corporate Life & Pensions
Jean-Michel Six
Standard & Poor's Ratings Services , Managing Director & Chief EMEA Economist
Philip Warland
Fidelity Worldwide Investment , Head of Public Policy

Objectives of the session

This plenary session was devoted to discussing the economic challenges and the impacts on existing pension systems posed by the ageing population. The session also addressed the contribution of the financial industry to these issues.

Executive Summary

1. The character of the problem


Significant changes in the age structure projected in the EU

In order to understand the nature of the problem, it was important to define what an ‘ageing population’ was. Here the source was low fertility rates, the retirement of the baby boom, and improvements in life expectancy. The other impact had been that the low fertility rate in most OECD countries was quite below the substation rate, meaning there would continue to be a reduction in the population. The primary factor, though, was improved life expectancy. On an individual level, people would spend longer in retirement than previously, but the number of years saving for retirement was not increasing.

This ageing affects the solvency of DB and DC funded pensions

The impact of population ageing and the micro and macro impacts differed depending on whether the subject was defined benefit (DB) pay-as-you-go pensions, DB funded pensions or defined contribution (DC) funded pensions. In DB systems the question was whether there would be enough money. If the ratio of people paying into the system moved to 4:1 to 2:1, then there would not be sufficient money, as pension benefits were paid with current contributions. In DC systems, the macro aspect was more important, as people would live longer and accumulate the same amount or less to finance longer in retirement.

The fiscal and economic challenges will be substantial in many Euro area countries

Ageing would mean that more people would withdraw from the labour force. Employment in the working population might decline, and pension expenditure was expected to remain the same until 2060. However, pension expenditure was only one element, alongside healthcare and long-term care, which was expected to grow by 2% by 2060. The current migrant crisis had benefits in the medium term, helping economies address severe ageing problems, but it also posed a challenged to EU governance. There was agreement about the need for people to work longer, to provide the income for later security, but that security needed protection in the present as well.

Demographic trends would impact the creditworthiness of sovereigns. Simulations showed that the US would reach sub-investment grade by 2050, the UK, France and Germany would be at BBB during most of the period. Since the simulations a number of countries had taken steps to address demographic change, by reforming their social security systems and reducing their budget deficits. However, real and nominal growth prospects had been lowered, as inflation had been all but absent. Median net general government debt in the advanced-market economies would be 134% of GDP by 2050, up from 52% currently.

The Regulatory Challenge and the political responses

Across Europe, questions needed to be asked as to whether entities were sufficiently solvent and able to deliver on promises. In that respect there was also pressure from the low-interest environment which diminished returns. If the situation remained unchanged, benefits would need to be reduced and people would need to work longer. There were two political responses: the populist, where politicians sought refuge in Pillar I, and the reformist, where better collaboration between the three pillars was sought. It was important for pension providers to be able to deal with consistent regulators and rules.

2. Public Awareness and Understanding of the Problem


Public Awareness

There was general agreement that people needed to work longer. Some people, however, lacked the proper understanding of changes being made in their own interests. Many of people’s beliefs about economics were untrue. Even when people understood, it did not mean that everybody acted appropriately. Similar to climate change or other hugely long-term challenges, there was a perception gap between long-term realities and short-term actions. Even if a person had the best intentions and started saving with a promise to save for the next 35 years, there might be incidents such as divorce, children or tragic events which caused the promise to be broken and the money to be taken out.

It was suggested that, while perhaps awareness was there on a conceptual level, it was not clear how many were fully aware of their own pension situation. Research had shown that most people thought the chance of their being unable to reach their pension without disability or a period of unemployment was smaller than 10%, when in reality it was 25%. In summary, understanding the problem was easy: people needed to work longer. The implementation, however, was less easy, and this was where the problem lay.

Government Awareness

The awareness was already present at a government level, as shown by the number of reforms which had passed in the last few years, such as automatic linking of benefits or retirement to life expectancy, or automatic stabilisers within the system. Three EU member states now had automatic balancing systems, eight had links between benefits and life expectancy, and seven had links between retirement age and life expectancy. As a result of this, predictions made only three years previously had proved pessimistic. Estimates suggested recent reforms in Italy would increase GDP by 2.5% and employment by 2.2% by 2030.

EU mechanisms had been introduced help governments to maintain awareness and to provide incentives for affecting those reforms, and the OECD message had always been to diversify sources in order to finance retirement. Private systems should be complementary to their public counterparts. In most countries, the solutions being pursued were to promote more and more DC schemes. The key thing was that benefits would be determined by the assets accumulated.

Addressing the Problems of Awareness and Understanding

Everybody recognised that life expectancy was increasing, but did they understand that they needed to save more? The savings opportunities and returns at present were not very exciting, owing to monetary and growth policies. One dimension so-far unmentioned was the rise in inequality. When these very complex issues were considered, a response had to be articulated that addressed the inequality aspect of recent trends. It was important to recognise that large parts of the European population would never be able to save for their pensions. It was important to help people to budget and allow them to save more.

With changes in the law in the UK, employers could no longer set a retirement age in pension schemes, so good pensions were needed to encourage employees to retire. Auto-enrolment had been introduced, which had been a very important first step.

3. The Solutions


Delaying Retirement and Greater Contributions

Although it had been said that people could not contribute more and contributing more led to lower returns, it seemed that lower returns resulted from higher life expectancy and lower growth. The only way to achieve what somebody who retired in the 1970s did was to contribute more and for longer. Many countries, such as Sweden, had done this by linking statutory age of retirement to improvements in mortality and life expectancy. This was fine, but care needed to be taken, as not everybody reaching retirement age had the same life expectancy. Many countries were linking statutory age of retirement to improvements in mortality and life expectancy, but care needed to be taken around socio-economic factors. The other suggestion was that the ratio of years contributing to years in retirement was important.

Increasing Participation

Increasing participation rates could address not only the ageing problem, but the growth problem more broadly. Improving participation rates of ages 55 to 64 could increase GDP by 3%, and, by improving participation rates of ages 65 to 75, the projected dependency ratio in 2060 fell from 2:1 to 3:1.

Increasing Productivity

In principle, you can borrow more, earn more or spend less to tackle such challenges. Borrowing more is, for the moment, out of the question, because debt is already very high, so we are left with either earning more or spending less. There, one can see that there is potential for spending less or at least spending better – there is potential for efficiency that could be tackled. Health costs can be reduced by 25% efficiency savings.

Then there is the better option, which is to earn more. Earning more with a lower population is more difficult and requires much more productivity. Improving productivity requires open markets within and outside the European Union. Entry costs and tax distortions which might create segmented markets would need to be removed. At the same time the quality of capital and labour needed to be improved, so more research and development was needed in innovation and more and better skills.

This was feasible, because, if one looked at the gap between the current situation and the best performing countries, it could be calculated that GDP could increase by around 11% in 20 years in the EU, though it varied from country to country. Additionally, if the reforms were implemented jointly by several or all member states, the dividend could increase to 12%. As a result, budget positions would improve, and so would employment. Once there was some sort of mechanism to remind governments of these benefits and entice them to make the reforms necessary to reach those dividends and communicate that information to the public. It was also important to create opportunities for older people to work on a part-time basis. This would require certain tax incentives and societal levies, to not disincentivise people working 10 to 15 hours from topping up their pensions.

Public/Private Co-operation

Pillar I would provide only a basic level of financial support for elderly people, and would need to be supported out of general taxation. Pillar II would need to be strengthened with more auto-enrolment, with a move to DC rather than DB pensions. This pillar was a combination of public requirements managed in a private fashion. Pillar III would then serve as more of a top-up to safeguard the standards of living for those who could not afford it.

In terms of auto-enrolment, the next step within Pillar II schemes would be auto-escalation. Employees received a 3% wage increase, 1% of which went into their pension. This could be done without legislation, and it was important that employers started to think about it.

Decumulation

Decumulation would also be more and more important. The insurance world could not provide the guarantees needed, so co-operation was needed. Following recent changes, the UK was a test-bed for examination of the different options. Mistakes would be made, but could be learnt from.

Other solutions

Other solutions included moves to transfer risk to the individual, with capital backed savings, so that people would contribute more. Financial markets had a very important role, because of the need for drawdown programmes and annuity markets, and insurance companies and pension funds both needed financial instruments to hedge longevity risk.

Detailed Summary

Introduction


Between 1970 and the early 21st century, the demographics in advanced-market economies, with the baby boom coming through, had led to a large increase in the labour force, causing a positive supply-side shock to the economy. In more recent years, this had been increased further by the entrance of labour forces in emerging-market economies, particularly China and Eastern Europe.

Demographic change was arguably responsible for many of the longer-term trends during this time period. Due to the increase in the labour supply, real wages had been stagnant for many years. There had been more workers than available jobs, inequality had risen as labour share decreased, and there had been low investment because profits were high and wages low. This had been profoundly disinflationary, and had allowed rates to stay very low for a long period of time.

The problem was that, in the early years of the present decade, everything was going into reverse, including the aforementioned trends. It was anticipated that real wages would rise in Europe and elsewhere, and perhaps more so in Europe, owing to the decline in the active labour proportion of the population being greater than in the United States. This would result in reduced inequality and higher investment, as well as an inflationary environment. Demographics had been a major driving force in the past (real per capita growth adjusted for population growth has been rather flat including in the economies with massive inflow of population, such as Spain in 2000s).

The effect for the EU would be that Pillar I would be very difficult for governments to honour. Governments also had a problem going forwards in terms of demographics. In addition to the government problem, many private-sector pension schemes were also in trouble. In many cases defined benefit (DB) schemes were underfunded, and defined contribution (DC) schemes might not generate sufficient income to provide older people with an acceptable standard of living. Both in the public and private sectors, demographics implied that there were problems coming down the line.

The character of the problem


The Demographic Challenge: dramatic changes in the age structure projected in the EU

In order to understand the nature of the problem, it was important to define what an ‘ageing population’ was. Everyone was aware that it was an increase in the number of old-aged people within the overall population. However, the source was more important. Here the source was low fertility rates, the retirement of the baby boom, and improvements in life expectancy. The baby boom was, in some sense, a temporary effect as once the baby boom cohorts passed away the bulge would disappear. The cost had already been incurred, so the solution, in some sense, was either to increase or reduce their benefits, with whatever impact it might have on the adequacy of their retirement income even to survive.

The other impact had been that the low fertility rate in most OECD countries, excepting Sweden and France, was quite below the substation rate. Therefore, there would continue to be a reduction in the population in the future. An important factor, however, was the improvement in life expectancy. Ageing had both a macro and micro impact. The macro impact was well-discussed: there would be an increase in the “dependency ratio” from 4:1 at present to 2:1. On an individual level, however, what was changing in terms of improvements to mortality and life expectancy was the ratio of the number of years people stayed in retirement to the number of years they saved for retirement. They would spend longer in retirement than previously, but the number of years saving for retirement was not increasing. This created a serious problem for the financial adequacy of those savings.

Ageing meant that more people would withdraw from the labour force, and potential growth was either declining or, as expected, hovering at around 1.5%. Employment in the working population might decline, and pension expenditure was expected to be around the same until 2060. Pension expenditure, however, was only part of the expenditure related to ageing, alongside healthcare and long-term care. All of this was expected to grow around 2% by 2060. An additional point was the migrant crisis currently confronting Europe, which was a huge challenge from a human, economic and governance standpoint. In the medium term, most experts recognised that there were potentially clear benefits to be obtained from the migration flow, which would help certain economies presented with particularly severe ageing problems to address them. Germany seemed to have, early on, recognised that this was an opportunity.

As always with human societies, opportunities also presented a challenge, as the migration crisis was challenging the EU and the governance of the Union itself. This was an additional risk that needed to be contemplated.

It was good that there was agreement about working longer, changing the pensionable age, and people being productive for a longer period. This would provide the income for their security later on. Another take was that the future security they were now going to finance also needed some protection now. What was sometimes forgotten was that increasing longevity had a problem called the ‘income protection gap’.

This ageing affects the solvency of Defined Benefits (DB) and defined contribution (DC) funded pensions

There was a need to distinguish between the different types of pension systems. The impact of population ageing and the micro and macro impacts differed depending on whether the subject was DB pay-as-you-go pensions, DB funded pensions, or DC funded pensions. DB pensions were based on a promise, and included the public pay-as-as-you-go system with current pensions paid by current contributions, and funded pensions, which were like DB pension plans. In terms of the first, which were referred to as Pillar I pay-as-you-go pensions or social security, the main problem was financial sustainability. The question was whether there would be enough money on the basis of unchanged policies. If the ratio of people paying into the system moved from 4:1 to 2:1 for each retiree, then there would not be sufficient money on the basis of unchanged policies, as pension benefits were paid with current contributions.

In DC systems benefits were determined by assets accumulated; however there would still potentially be a problem of inadequacy. The more important here was the macro aspect, as people would live longer and accumulate the same or less to finance longer in retirement, creating a problem.

Clearly there was a challenge, and improvements were needed to all three pillars for a satisfactory solution. The opportunity, though, was that some countries were more advanced in respect to some pillars rather than others, and therefore there was room for catching up on best practice. Every OECD and European country had the three pension types, and the only difference was the weighting.

The fiscal challenge will be substantial in many Euro area countries

Among other challenges, demographic trends were on the basis of unchanged policies likely to significantly impact the creditworthiness of sovereigns in the long term. It had been recognised a long time previously that ageing, along with global warming, was amongst the most important issues over the next few decades, particularly for developed economies. A rating agency had for more than 10 years been publishing assessment on what the impact might be on creditworthiness of sovereigns, assuming an absence of specific policy actions to modify the impact of demographic changes on budget trajectories.

In their latest study, under a no-policy-change scenario more than one-quarter of the 58 sovereigns analyzed would by 2050 have credit metrics that are currently associated with speculative-grade sovereign credit ratings ('BB+' or below), against less than 10% of this sample in 2020. Simulations had showed that the US sovereign, currently at AA+, would fall to sub-investment grade by 2050. The UK, currently AAA, would be at BBB level throughout the same period, as well as France and Germany.

The simulations results compare favourably to previous vintages of the study. On the one hand, it was clear that a number of countries had taken brave decisions to begin to address the demographic challenge through modifying their pensions systems, by delaying the age of retirement or monitoring certain contributions, as well as health-care systems. At the same time, in most sovereigns budgetary consolidation advanced and reduced the budget deficits. On the other hand, growth prospects had been lowered, in real GDP terms, but also in nominal terms, as inflation had been all but absent. The change in growth prospects in the last 10 years was an aspect that needed to be underlined.

Without any policy change, the median net general government debt of advanced economies will be 1343% of GDP by 2050, but in emerging markets it will reach even higher levels at 136% and from a lower starting point (respectively, 51% and 42% of GDP in 2015).

Failure to ensure adequate level of pensions could constrain access to health-care services even as the demand of senior populations for them increases. Policies will thus have to address health care spending to ensure sustainability, while also preventing increases in poverty risks for the elderly and social inequality. In this context, we observe that rationalizing social security systems can, if embraced early on, spread the impact and the burden of unpopular policy measures.

For example, in the UK one third of households fell back one quintile in household income when the main income earner was disabled or ill for a long period. Perhaps more shockingly, 20% of UK households fell into what might be called poverty. The consequences were not only at the household level; people would have even less money to save for their pensions, and would dip into their savings to survive on a daily basis. This was the reality now in the UK and other markets. For governments this had severe consequences, as they would have to help out one way or another, as the money could not be used for making pots available for social security or pensions.

The Regulatory Challenge

There was also a regulatory challenge. Looking at what was already in existence across Europe, the questions that needed to be asked were whether such entities were solvent enough, and whether they could deliver the promise, both politically and economically. In that regard there was huge pressure on two fronts. It was not just the already-discussed demographic trends, but the low-interest environment which drastically diminished the returns upon which the financial systems and safeguards were based. The combination of these two aspects, in the medium-to-long term rather than the short term, would create devastating results if it continued. There was no way around it; if the situation remained unchanged, benefits would, one way or another, need to be reduced and people would need to work longer.

The political challenge

This led onto another challenge: the political one. This split into two fundamentally different positions which were currently visible. The first was the populist response, where politicians sought refuge in the Pillar I pay-as-you-go world, as they saw Pillars II and III as disappointing. Thus they returned to Pillar I, which had proven stable over the previous 50 years. This was common across the left and the right, which applied a populist position that was completely economically unsustainable. Political pressure from the electorate to do something which, in the short term at least, promised some safeguard was massive.

The second prong of the political challenge was more of a reformist agenda, which had been understood and needed to be tackled. At the end of the day they needed to find a better calibration of the interaction between the three pillars. Any concept based on just one pillar was going to fail. The combination would need to be very smart, as all countries had the three pillars in one way or another, but in different compositions. The Netherlands was the leading country in Europe in regard to occupational pensions, but that case was based on 50 years of prudent policy. It was not as easy as pushing a button and making the change in a day. Many countries, including Germany, would need to increase enrolment in Pillar II, but that would not be an easy answer to these challenges. There would be painful remainders which would need to be resolved differently.

The public and private sectors needed to start to collaborate and go beyond words. For pension providers, it was important to be able to deal with consistent regulators, but in the UK budget new rules were introduced every time. The political drive behind it was understandable, but it would not help.

Public Awareness and Understanding of the Problem


Public Awareness

There was an understanding of the implications of an ageing population. People were living longer, so had to finance more. The promises had been made, so either they would be cut or not fulfilled. At the end of the day, even adding the environment of low growth, low wages, low returns and low interest rates, the only solution was to contribute more for longer. Attempts had been made to deal with this via postponing the official or statutory retirement age, but care was needed as the data showed that increases in life expectancy differed between socioeconomic groups. Therefore, increasing the retirement age could be problematic. There was general agreement that people needed to work longer, taking into account that depending on socioeconomic features or characteristics such as education, income and different occupations, improvements in and levels of life expectancy were quite different.
Some people did not seem to have a proper understanding of changes being made in their own interests. This was a much broader occurrence, and a number of papers had been written about it, most recently False Beliefs and Unhappy Endings. Many of people’s beliefs about economics were untrue.

Was the problem being tackled? Yes, and no. Everybody understood the problem, both on the public and private sides, even citizens, but this did not mean that everybody acted appropriately. There were hugely conflicting agendas, which did not suggest a single, obvious solution, as the solutions that were needed were not easily conveyed to a broader electorate today. It was curious that younger people in France, who would benefit most from changes, resisted any change where there was a perception problem. As with climate change or other hugely long-term challenges, there was a perception gap between long-term realities and short-term actions.

The biggest threat of poverty in old age was not to well-off or well-earning middle class people, but to borderline and low-income households who could not save a lot anyway, or those working in SME-type companies which did not provide massive occupational pensions. Another driver behind the need for a higher and different level of mandatoriness was commonly called ‘broken CVs’. Even if a person had the best intentions and started saving with a promise to save for the next 35 years, there might be incidents such as divorce, children or tragic events, which caused the promise to be broken and the money to be taken out.

It was suggested that, while perhaps awareness was there on a conceptual level, or among those who operated in this world, it was not clear how many were fully aware of their own pension situation, including the gap they might have and how much money they might need. This question was often asked when they were developing a new pension product. It was shocking how little people knew.

Research had been done with Oxford University, where people were asked what they felt was the chance they would be unable to reach their pension without short- or long-term disability or a period of unemployment. More than half thought the chance was smaller than 10%. The real number was 25%. This was just a statistic, but could have severe consequences for households.

It was also not clear to what degree communication would help, as it could not be looked at from the top or the bottom, and it was not very interesting. A 16-year-old would never want to discuss pension provision. Auto-enrolment combined with tax incentives might not be unavoidable, but it was something that required more research. Again, this could not be done without strong collaboration between public and private sectors, and some communication. More time needed to be put into looking at this as an industry and a public sector.

Given the reception that rating agency’s reports and simulations had received from policymakers and public opinion in general, it was suggested that the awareness of the problem had increased significantly over the last few years. Policymakers and the public in particular, had become much more aware of the severity of the problem represented by ageing societies.

In summary, understanding the problem was easy: people needed to work longer. The implementation, however, was less easy, and this was where the problem lay.

Government Awareness

The awareness was already present at a government level, and while it was less present at a public level, there was some awareness. The proof was in a number of reforms which had passed in the last few years, touching on various aspects. One was the automatic linking of benefits or retirement to life expectancy, or automatic stabilisers within the system. This had been passed in a number of countries: three EU member states now had automatic balancing systems, eight had links between benefits and life expectancy, and seven had links between retirement age and life expectancy.

As a consequence, predictions made only three years previously had proved pessimistic. Three years previously, it had been expected that by 2060 total expenditure would increase by 3.5% of GDP, whereas the latest estimate was that it had fallen to 2%, so there was some impact in the long run. The average exit rate from labour participation had also improved in the latest reforms. Some conservative estimates of the recent measures in some member states showed that, for example, recent reforms in Italy would increase GDP by 2.5% and employment by 2.2% by 2030. This fed into awareness and possibilities of making further progress.

This was accompanied by EU mechanisms to help surveillance and to help governments to maintain awareness and to provide incentives for effecting those reforms. There were, then, a number of mechanisms and measures taken at EU level that kept alive the challenge. Every three years there was an ageing report, and every year a fiscal sustainability report published by the EU Commission. There was also an annual peer review at Council level, so as to really push member states to take the required action and to explain to their populations that this was the best, if not the only, way forward.

Another point was around improving co-operation between the three pillars. The OECD message had always been to diversify sources in order to finance retirement. Public and private were needed, as were DB and DC. Additionally, the private system should be complementary to the public system. The political setup would determine the combination, but it was important to understand how the combination defined things.

In terms of solving the problem, the solutions that were being implemented in many countries were in addition to the pension system reforms, included a promotion of DC schemes. Whether they were Pillar II or III, the key thing was that benefits would be determined by the assets accumulated and the financial market performance. This required financial products for people when they reached retirement age, as people needed to mix drawdown programmes with annuities that provided protection from longevity risk. Then annuity markets were needed, and even in the UK – the most developed annuity market – these were fraught with problems. Insurance companies and pension funds needed financial instruments to hedge longevity risk; otherwise they would not be able to offer such products.

Approaches in Different Countries

There were a couple of things that were already happening and needed to be further addressed. In Germany, as was widely known, very severe reforms had been employed eight to 10 years ago, which in hindsight had proven to be extremely beneficial with regards to pension systems. In the past two or three years, quite a bit of that had been gambled away. Even once parts of the problem had been tackled with pretty bold reforms, it did not mean that they would stay addressed for the next 30 to 50 years, because the fundamental and underlying trends were not going away, and had not reversed.

A couple of key components would need to be considered in order to honestly tackle the problem. The first was that regardless of whether people liked it, more risk would move to the individual. Secondly, pay-as-you-go systems would not be the solution in terms of a component to be increased. They had to be part of it, but could not be the solution to all of the problems. Thirdly, capital-backed savings needed to increase, but not necessarily through the private sector alone. There were schemes and ways in which to organise capital-backed that were not necessarily left to the private sector alone in a traditional Pillar III private pension sense. If the aim was really to make a change, then this would probably need to be combined in a smart way with capital-backed savings with a higher level of mandatoriness. The key would be how to get this done.

What was needed was an intelligent combination of private savings and public subsidies in a smartly-calibrated, semi-mandatory sense. Otherwise, telling people to save more would not work. This was not how elections were won, and that was part of the problem.

In France, they had increased the age of retirement, but had allowed people to retire at 62. The French public system was different to many others in the OECD, as it linked the number of years of contribution required for eligibility for a full pension with improvements in life expectancy and mortality. Retirement at 62 in France was only possible if a person had 41.5 years of contributions into the public pension system, and this length of contributions increased with improvements in mortality and life expectancy.

The French system was based on the macro system, which was why it was beginning to separate out the different aspects. It was based on the number of years contributed, relative to the number of years in retirement, having to remain constant. This contribution period was linked to life expectancy. What other countries had done was to link the statutory age of retirement to life expectancy, which applied to everybody across the board. Those who started working at 16 and reached 67 had 40-odd years of contributions, and those who started work after their PhD at around 30 would be entitled to a full pension after 35 years.

Addressing the Problems of Awareness and Understanding

The real issue, though, was around disputes and uncertainties as to what could and what should be done about it. Everybody recognised that life expectancy was increasing, but did they understand that they needed to save more? Maybe they did not, and maybe they should save differently, rather than more, but this was very deflationary. If a society saved much more than it had used to do, what sort of domestic demand would be the result, and where would it be saved? The savings opportunities and returns at present were not very exciting, owing to monetary and growth policies.

Here there were still wide disagreements between economists, and uncertainties around what else, besides the previously mentioned policy measures, should be done to address the issue, besides making more babies. That was one option, but not the sole option.

One dimension which had not been mentioned so far but which complicated the public debate was the rise in inequality. This had been widely documented, especially by the OECD, over the last 10 years. In developed economies there had been a significant increase in inequality. When policymakers suggested new measures and reforms to address the ageing issue, it was frequently perceived by public opinion that it would harm those hardworking components of the population, while those at the top would be unaffected; they could not care less what age they retired at, because they already had lots of money. This had been widely exploited by populist parties in various countries. When those very complex issues were considered, a response had to be articulated that addressed the inequality aspect of recent trends.

The debate had to recognise that large parts of the European population would never be able to save for their pensions. This put pressure on Pillar I, as it could only offer falling real incomes over the long term due to the fiscal straitjacket. It was not just people who could not afford a single euro; in the United States an asset manager had the largest provision for 401(k)s. There customers had requested that they stopped telling them to save more. When queried why, they said that they had no money. As a result, they had set up a site to help people budget, and to allow them to save more. It was not that people had nothing to save, but was that people needed to know how to manage their finances.

One of the things this asset manager had been finding in the UK with employers for whom they ran pension schemes was that they were no longer able to set a retirement age and force employees to retire. Thus they wanted good pension schemes so that their employees could retire. They did not want employees going on and on, as it caused succession problems.

Auto-enrolment in the UK had been produced after the Turner report, which had essentially been a cross-party report. The proposal had been made by a Labour government, initiated by a coalition government, and followed through by a Conservative government. It did not force people to save enough, but it had been a first and very important step along the way.

The Solutions


Delaying Retirement and Greater Contributions

There was a need to distinguish between different sources. In terms of the baby boom, the large cohorts born in the 1950s and 60s were now going into retirement and could not save any more. Debt could be increased to pay the promise, but was the promise going to be kept? If it was cut, would people complain? Some average workers in most countries who were close to retirement had been told what they needed to do in their working life, and what they would receive, but suddenly they were reaching retirement and their benefits were going to be cut. It was a tough decision to make, but it was one that had to be made. In terms of the baby boom, there was not much else to be done.

Going forward, in terms of improvements in life expectancy, reforms could be made to the pay-as-you-go public systems as well as to systems in which promises or benefits were backed up by capital. The main reform was around what had already been discussed: working longer and contributing more. Although it had been said that people could not contribute more and contributing more led to lower returns, it seemed that lower returns resulted from higher life expectancy and lower growth. The only way to achieve what somebody who retired in the 1970s did was to contribute more and for longer.

The key question was how to implement postponing retirement. Many countries, such as Sweden, had done it by linking statutory age of retirement to improvements in mortality and life expectancy. This was fine, but care needed to be taken, as not everybody reaching retirement age had the same life expectancy. Socioeconomic factors were very important. Some people start working earlier than others, so contributed to the system, in particular the public system, for much longer. Looking at the numbers, certain groups contributed for longer than others, due to having started working earlier.

The other option went back to the idea that it was the ratio of years contributing to the years in retirement that was important, which was the idea of a DC scheme. The benefits were perfectly linked to assets that had been accumulating during the working life, and contribution periods could be linked to improvements in life expectancy and mortality.

Increasing Participation

In terms of low participation, female participation, low-skill participation, and old-age participation could be increased. This was not only to address the ageing problem, but to address the growth problem more broadly.

In terms of participation rates, it was estimated that improving the participation of ages 55 to 64 would increase GDP by 3% in the long run. Looking at the 65 to 75 age group, as mentioned earlier, the dependency rate could fall from 4:1 to 2:1 by 2060. If participation rates of the 65 to 75 age group were taken into account also, this fell to 3:1, which was a significant improvement.

Increasing Productivity

In terms of how to tackle that, there are three ways. In principle, if you fly on automatic pilot, you can borrow more, earn more or spend less. Borrowing more is, for the moment, out of the question, because debt is already very high, so we are left with either earning more or spending less. There, one can see that there is potential for spending less or at least spending better – there is potential for efficiency that could be tackled. Health costs can be reduced by 25% through efficiency savings.

Then there is the better option, which is to earn more. Earning more with a lower population is more difficult and requires much more productivity. Improving productivity requires open markets within and outside the European Union. Entry costs and tax distortions which might create segmented markets would need to be removed. At the same time the quality of capital and labour needed to be improved, so more research and development was needed in innovation and more and better skills.

This was feasible, because, if one looked at the gap between the current situation and the best performing countries, it could be calculated that GDP could increase by around 11% in 20 years in the EU, though it varied from country to country. Additionally, if the reforms were implemented jointly by several or all member states, the dividend could increase to 12%. As a result, budget positions would improve, and so would employment. The EU has governance mechanisms to remind governments of these benefits and entice them to make the reforms necessary to reach those dividends and communicate that information to the public.

In addition to the more technical elements of the traditional three pillars, it was very important to offer and create more opportunities for older people to work on a part-time basis. Simply increasing productivity in the general age brackets, which was important, would not be sufficient. Many would depend on an ability to work part-time later on, which would require a certain set of tax incentives as well as social levies in order not to disincentivise people to put in 10 to 15 hours’ worth of work per week to top up the pensions they already received. It was a mixture of those tools which would, hopefully, create a sustainable public policy answer going forward.

Sometimes there was a bit of an oversight of the so-called ‘silver economy’, which also presented new opportunities for jobs. Specifically, in an ageing society demand would emerge for new types of services that responded to the demographic challenge. In particular, this would mean services for senior citizens, etc. It was not clear that much thinking had been dedicated to how to address those needs and how to train people to work in this relatively new dimension of society and the economy. This was not only in the medical sphere, but in a broader range of services that would increasingly be in demand. It was not clear that the education system was already contemplating how to address these needs and train people to work full-time along those dimensions.

Public/Private Co-operation

The first pillar, a pay-as-you-go system, was likely to only provide a very basic level of financial support for elderly people, and even this would increasingly be supported by general taxation, and less by specific levies. Those levies would remain, but the share of general taxation would increase, which was not such a bad thing from a social policy perspective, as it addressed some of the previously-mentioned challenges. It was more the general society support which went into that system.

The second pillar, occupational pensions, would definitely have to be strengthened via more auto-enrolment. The future would belong to DC rather than DB. A very smart mixture of tax and cash subsidies in that space, particularly for low-income earners, would be very important, in the occupational space as well as the private-pension space. That second pillar, of course, could be run by private enterprises. It was a combination of public requirements or constraints managed in a private fashion.

The third pillar would serve more as top-up for those who could not afford it, in order to safeguard their personal standard of living, using a wide variety of tools that were already available.

One of the most hopeful things was the work being done by the European Insurance and Occupational Pensions Authority (EIOPA) on the Pan-European Personal Pension (PEPP). Gabriel Bernardino had written a very elegant and eloquent piece about the PEPP or something like it in Pillar III.

In terms of auto-enrolment, the next step within Pillar II schemes would be auto-escalation, which had worked well in the United States. Employees received a 3% wage increase, 1% of which went into their pension. The difference was unnoticed, and backed up by the employer. This could be done without legislation, and it was important that employers started to think about it.

Decumulation would become more and more important

The insurance world would not have sufficient capital to provide the guarantees the world wanted, so other collaboration would be needed. An asset manager had just collaborated with an insurance company, where they did the underlying asset management and tried to achieve 4% forever. They could not possibly guarantee it as an asset manager, but the insurance company could, as it was their product. This was the sort of innovation which was needed.

On decumulation, the advice was to watch the UK. Following changes there, the UK was now a laboratory and test bed where people could take the money, buy an annuity, drawdown or do nothing. It would be interesting to see how people behaved in such an environment. Mistakes would be made, but they could be learnt from.

When public and private sector providers and regulators worked together, solutions like the PEPP as well as new longevity products could help. On the latter, it was not only about longevity risk but also about the investment risk being managed. Currently there would not be sufficient capital around, though this sounded counterintuitive. The collaboration and innovative thinking at this meeting would have to finance the solution, as there was no other choice.

Other solutions

Moves to transfer risk to the individual, which were being done much more in Latin America and Asia than in Europe, combined with capital-backed savings, were essential, as was the issue of introducing incentives, as this was the only way that people would be made to contribute more.

This was all happening in the world of funded DCs, where people reached retirement and had an amount of assets accumulated to allocate to finance their retirement. Financial markets had a very important role, because of the need for drawdown programmes and annuity markets. Annuity markets did not work perfectly, and insurance companies and pension funds needed financial instruments to hedge longevity risk. There was a need to distinguish between different kinds of longevity risk: idiosyncratic longevity risk and aggregate longevity risk.

Conclusion


People were saying that the situation was unsustainable, and if it was unsustainable then it would stop. The question was whether it would stop in an orderly or disorderly way. The purpose of the exercise at hand was to make sure the solution was orderly.

They had discussed demographics as a problem, and the framework that had been used in terms of a solution was a production function: output depended on labour inputs, capital inputs, and total factor productivity, all of which should be improved. There was a huge on-going debate about the future of productivity.

On the one hand were people who said that all of the good inventions had already been invented, and productivity would be ‘in the doldrums’ for decades. On the other hand were people who said that, given the way forms of learning in many disciplines could be combined, such as in the iPhone, an incredible increase in productivity was about to happen. In the latter case, no problem such as had been discussed would happen, but there would be a huge problem of inequality if it was only some people who received the benefit. A German expression held that, ‘For every solution, there is a problem.’

Secondly, it had been suggested that if they were not careful more saving would mean less demand. The paradox of thrift meant that, when everyone was trying to save more, in the end they saved less. If there was more saving without more demand through investment, there would be a problem.