What Makes Pension Reforms Sustainable ?

Policymakers pushing pension reforms have tended to justify changes on the basis that they would make systems more sustainable by lowering future spending on pensions. This is a rather narrow interpretation of sustainability that fails to consider that other fiscal programs may need to accommodate the impact of reforms that reduce pension system adequacy. In this light, this article argues that in order to correctly assess the sustainability of pension reforms, one needs to adopt a more holistic framework that encapsulates the interaction between pension system goals and constraints. In a number of countries, reforms focused solely on reducing future spending were followed by reforms that restored generosity. A holistic approach to assess pension sustainability could help limit this cycle of reform and increase trust in pension systems.


Introduction
State pension systems are the cornerstones of modern welfare states in many senses.Holzmann and Hinz [1] portrayed their rise as a reaction to the societal and economic changes of the nineteenth century, with the informal arrangements that had existed in traditional agricultural communities being replaced by arrangements that enabled industrial workers to smooth their income over their lifetime and prevent poverty in old age.The first state income-transfer programs towards the elderly can be traced to the late nineteenth century in Germany and Denmark [2].In each of these countries, the emphasis of the system differed.In Germany, the main focus was on ensuring that after reaching a certain age, workers could stop working and still have an income that helped them maintain their previous living standards.In Denmark, the program was a means-tested scheme for poor citizens aged over 60, with a clear emphasis on poverty alleviation.Barr and Diamond [3] demonstrated that income security in old age requires two instruments: a means to facilitate consumption-smoothing over the life course and a means of insurance against the risk of old age poverty.The distinction in the importance given to these two system goals has led to the rise of an extensive literature with several taxonomies of pension systems, as described in Soede et al. [4].
Up to the 1970s, state pension systems were on the rise across the world.Only a small number of mainly American economists, such as Feldstein [5], raised dissenting voices, and the concern tended to be on whether state pension systems displaced private saving and lowered long-term economic growth rather than on the size and scope of retirement income provision.All this changed in the 1980s mainly due to two factors.On the one hand, the stagflation of the 1970s and the sharp international recession of the 1980s had led to ballooning government deficits and significant balance of payments crisis in many economies.Given that pensions had become either the largest or one of the largest government spending programs in most countries, reforming them was seen as a way to address these internal and external imbalances.On the other hand, policymakers woke up to the largely unanticipated spectacular rise in longevity and decline in fertility that had characterized previous decades.For instance, despite the long and respected actuarial tradition on which they were based, its constraints.For the cycle to stabilize and pension reforms to prove more sustainable, a holistic approach needs to put in place whereby reformers focus on both system goals and constraints.
This approach is quite different from path dependence, as espoused in Pierson [21] as the latter implies that history matters but in a way that changes in systems become ever more incremental and follow the same direction.Subsequent literature, such as Andersen and Larsen [22] and Hinrichs and Kangas [23] modified the static path dependence theory, introducing the concept of a martingale motion, whereby a system departs from its previous past through incremental changes which appear small at first but which eventually become irreversible and self-reinforcing.Conversely, Sefton, van de Ven and Weale [24] argued that as the age profile of voters in a country becomes older, the tendency will be to limit reforms in countries with generous systems while in those with more residual systems, there would be pressure to improve benefits.
This article argues that the arguments on path dependency focus too much on institutional arrangements, and that rather the reform process is constrained by the objectives, or expectations, of the public and that whenever constraints force changes that reduce the latter, it is only a matter of time before there is pressure for policymakers to restore a similar balance as before.As a result, the cycle of continuous pension reform can only change if expectations are changed substantially or if the nature of the constraints faced by the system changes.The recent evolution of pension reforms suggests that citizens appear to be accepting a smaller role for income smoothing through the pension system, while accepting that longevity requires longer effective working lives.The impact of these changes on system goals and constraints could mean that the pace of the cycle of reforms could slow down, but unless future reforms adopt a more holistic approach to assess pension system sustainability, it is likely that the process of to-and-fro will continue.

Pension Reforms-Cyclical Tendency or Secular Trend
The policy discourse on pensions since the 1980s has tended to be dominated by a narrative of crisis both in academic economic circles, via Disney [25], and in policy institutions, via the World Bank [2].For pensions to remain 'sustainable', only one reform direction-that of retrenchment-was envisaged.This reflected the very negative change in opinion on the effectiveness of the pay-as-you-go approach to financing pensions.The latter approach, which previous economists such as Samuelson [26] had championed, was increasingly depicted as a sort of Ponzi scheme whereby earlier generations of workers had lumbered future generations with a debt burden for expenditure outlays they would not benefit from.Some economists, like Feldstein [5], also argued that the introduction of pay-as-you-go had contributed to lower long-term economic growth as it led to reduced saving and, in turn, to lower private investment.Since the ageing transition, in particular the retirement of the Baby Boomer generation, was seen as inducing lower economic growth and reducing the relative size of the contributor population, keeping pension policy unchanged was portrayed as a recipe for disaster.In fact, substantial pension reforms ended up being one of the main policy recommendations of international institutions, particularly moving emerging countries away from the 'mistakes' of retirement income provision systems prevalent in the developed world.The Independent Evaluation Group [27] indicated that the World Bank, for instance, assisted 68 countries with reform of their pension systems with more than 200 loans and other forms of support between 1984 and 2004, with greater resources granted to countries developing multi-pillar systems.
The sustainability narrative underpinning the pension reform process was firmly rooted in the concept of sustainability used in public economics.In this branch of economics, the theoretical framework is based on a representative agent model where besides fulfilling an intertemporal budget constraint, government also has to constantly fulfill a static budget constraint.This approach, for instance adopted in Chalk and Hemming [28], implies that to be solvent the present value of future surpluses must be higher than that of future deficits so that all government debt is repaid.Caldarelli, Sefton and Kotlikoff [29], in fact, argued that the government's intertemporal budget constraint meant that any bills left unpaid by current generations, such as the implicit pension debt of public pension systems, must be paid by future generations.
Hauner, Leigh and Skaarup [30] presented the standard fiscal sustainability assessment of pension policies made by international institutions-an approach that underpins for instance the assessments of national policies made by the European Commission.The authors assess fiscal sustainability through the 'debt target primary gap' and the 'intertemporal primary gap'.The first measure computes the fiscal adjustment required to achieve a given level of gross public debt-to-GDP by a certain year.The second measure computes the fiscal position required so that public debt remains unchanged over time.Given the projected expansion of pension spending and long-term reduction in economic growth, combined with the large initial level of government debt, the policy advice emanating from these models is retrenchment of the public pension system.For instance, Annicchiarico and Giammarioli [31] pointed out that reducing benefits rather than raising taxes has a positive impact on economic growth and accelerates the speed of debt adjustment towards this target.
International institutions have frequently pointed out public pensions as being a fundamental stumbling block towards achieving long-term fiscal sustainability.In October 2006, the European Commission [32] published a communication to the European Council and the European Parliament where it noted that "in the coming decades, the size and age-structure of Europe's population will undergo dramatic demographic changes . . .this will make it difficult for Member States to maintain sound and sustainable public finances in the long-term".Similar statements have been made by entities like the World Bank, the International Monetary Fund and the Organization for Economic Co-operation and Development.As a result, assessments of fiscal spending have tended to be the primary measure of sustainability adopted by policymakers and researchers.Schneider [33], for example, determines success in pension reform on the basis of the extent of the decrease in expected spending on public pensions by 2050.
The European Commission, to give another practical example, carries out long-term forecasts of public spending on pensions across all Member States every three years and these are used as an indicator in its annual economic assessments of national policies.The definition of the indicator is particularly important as it focuses on the change in pension spending as a percentage of the gross domestic product over a period of time.This does not take into consideration further factors such as the initial level of spending, whether the change is in line with demographic developments, or if it reflects a slowdown in the gross domestic product rather than an acceleration in spending.
Howse [34] argued most governments are constrained when they are reforming pensions by the belief that spending is already too high, thus increasing taxes or public borrowing to fund even higher future spending is not an option.In this frame of mind, reducing future public spending, by carving out part of the public scheme and pushing new workers into private schemes would make the system more sustainable, as the size of future pension entitlements decline.The catch, of course, is that spending for current generations of pensioners cannot be financed from contributions of current workers, as these instead go into private pension schemes.When the financial crisis hit public finances, this catch made many countries that had followed the World Bank advice in the 1990s and 2000s regret their decision and revert back to pay-as-you-go financing.
While the case of the to-and-fro of Eastern European countries from pay-as-you-go pension systems is interesting, it does not necessarily undermine the Pierson hypothesis of a secular trend of welfare austerity.As a matter of fact, one could see it as a vindication of the path dependency theory.Eastern European countries tried to move away from their longstanding institutional frameworks, and at the first economic shock they were forced back.However, this would not mean a change from the secular trend of reducing pension system generosity.In fact, many reformers in these countries had sold the move to personal pensions as a way of ensuring that future pensioners would have better benefits, as returns on private pensions (returns on financial assets) were seen as higher than the return they would get from the pay-as-you-go system (the economic growth rate).The financial crisis, on the other hand, confirmed the theoretical arguments against this supposed benefit made by Orszag and Stiglitz [35] and Barr [36] nearly a decade before.Moreover the financial crisis showed that contrary to what had been conceived at the World Bank [2] when pressed for funds, many governments opted to confiscate private pensions as a more politically acceptable option than reducing government spending.
To argue for a cycle in pension reforms, one requires more compelling cases: where the key changes are not institutional arrangements but rather pressures arising directly from the effects of past pension reforms.This raises a complication as most countries have adopted a very incremental approach to pension reforms.Bonoli and Palier [37] in their review of reforms in European Union countries, noted four stages of reform.Until the late 1980s, the only changes were limited to raising contributions to finance funding shortfalls.This was followed by some moderate cuts in generosity; particularly changes in indexation of benefits.While relatively minor, these reforms brought the future of social security into the public debate.More radical reforms were conducted in the early 1990s, but reforms were still not one-sided in terms of retrenchment.This stage also saw the first moves towards funded private provision.Finally, the reforms conducted after the late 1990s were more radical with a stronger push in favor of private pension funds, lowering of benefits, measures to stop early retirement and increases in pension ages.
The authors emphasized that the more substantial reforms carried out since the late 1990s have long phase-in periods, which ensure the large and increasing politically influential cohorts of baby boomers, who will be retiring over the next two decades, will only be marginally affected by reforms.For instance, the authors showed that only about one in seven of the current Italian electorate are affected considerably by the major pension reforms.An added complication is that in some countries which went for significant systemic reform, major elements of the reform, such as the reduction in benefits induced by higher longevity, have been implemented haphazardly, according to Gronchi and Nistico [38], or have been delayed in the aftermath of the financial crisis, as shown in Grech [11].
This leaves one with very few examples of countries where the pensioner population is already feeling the effects of reforms and where there was pressure for change.The prime example is the United Kingdom, where one can clearly see a recurring cycle of reform spurred by changes in pension system outcomes.It is not the purpose of this article to describe in great detail the features of the pension system of Great Britain and its reform over time, as there are ample studies on this topic, such as Bozio, Crawford and Tetlow [39] and Pemberton, Thane, and Whiteside [40].
Figure 1 attempts to summarize the main changes that have occurred in the past fifty years.These are grouped into three different sets of reforms.Starting from the left, one has the reforms conducted in the 1970s, which aimed at improving generosity and enhance the coverage of the system.Arguing against their cost, reformers in the 1980s undid these reforms, while in the 2000/10s when the impact of the 1980s reforms became apparent, the latter were also undone returning the system to something more in line with the concerns that had inspired the 1970s reforms.
In the 1970s, the British pension system was still going through its expansion phase with concerns about its adequacy and the extent to which it was ensuring a good standard of living for pensioners.The two main reforms were the introduction of indexation to earnings growth of the value of the Basic State pension (BSP)-a flat-rate pension-whose value is determined by the number of contribution years paid, and the introduction of the State Earnings-related pension (SERPS)-which contrary to BSP was linked to the preretirement earnings of the contributor.In the 1980s, the Conservative government moved the indexation of the BSP to inflation, while it pushed for the abolition of SERPS.When this move met with political opposition, it instead introduced contracting-out, which meant that an individual could forgo paying part of social security contributions if they instead contributed to a private pension, either occupational or personal.
The impact of the 1980s reforms took time to be felt, but by the late 1990s the impact led to considerable political pressure on the part of pensioners.McGuinness [41] showed that in the 1960s and 1970s, the proportion of pensioners in relative low income had hovered between 35% and 40%.By the mid-1980s, it had fallen to closer to 15% showing the positive impact of the pension reforms of the 1970s.All of this changed in the late 1980s such that by the early 1990s the proportion was back to over 40%.This required significant intervention primarily through non-pension welfare benefits such as income support, which gradually lowered the poverty rate.However, by the mid-1990s the poverty rate among the UK's pensioners was still above 30% or twice that among German pensioners or four times that among Dutch ones.At the same time, contracting out had turned into a pension mis-selling scandal with even the largest UK insurance firms (Bennett and Gabriel [42]) implicated in schemes that led to worse pension outcomes for individuals.The resulting lack of trust, combined with subsequent heavier regulation of the pension sector, meant that the take-up of private pensions nose-dived.It is quite poignant that despite these two developments-namely very high pensioner poverty and the decline in private pension saving-many UK economists (for instance, Dilnot, Disney, Johnson and Whitehouse [43]) still thought their pension system was in a better state than other systems in Europe, on account of its low level of spending and that improving the state pension was not an option.
Sustainability 2018, 10, x FOR PEER REVIEW 6 of 12 pension mis-selling scandal with even the largest UK insurance firms (Bennett and Gabriel [42]) implicated in schemes that led to worse pension outcomes for individuals.The resulting lack of trust, combined with subsequent heavier regulation of the pension sector, meant that the take-up of private pensions nose-dived.It is quite poignant that despite these two developments-namely very high pensioner poverty and the decline in private pension saving-many UK economists (for instance, Dilnot, Disney, Johnson and Whitehouse [43]) still thought their pension system was in a better state than other systems in Europe, on account of its low level of spending and that improving the state pension was not an option.The Labour administration that came in power in the late 1990s, on the other hand, took a diametrically opposite view and conducted considerable reforms, which are amply described in Hills and Stewart [44].Besides introducing a more generous system of means-tested benefits, known as Pension Credit and a more progressive State Second pension (S2P), it set up an independent Pensions Commission widely considered as one of the major policy successes of the UK government (see Rutter, Marshall and Sims [45]).This Commission advocated the joining up of the flat-rate and earnings-related state pensions into one more generous benefit, the reintroduction of earnings indexation of state pensions and the auto-enrolment of workers into workplace pensions.The latter included the UK government setting up a low-cost workplace pension scheme so that not all employers had to create their own scheme, and also to act implicitly as a competitor to private schemes.The impact of the Labour administration's pension policies has been very pronounced, such that the pensioner poverty rate nearly halved and is now at par with that in Germany.
The subsequent Coalition and Conservative administrations continued these policies, in contrast to changes effected in other social welfare policies, such as working-age benefits (more details can be found in Lupton, Hills, Stewart, Burchardt and Vizard [46]).The Coalition administration even enhanced further the indexation of state pensions to include a triple lock that ensures a minimum increase of 2.5% even if both earnings growth and inflation are below this rate.The Department for Work and Pensions [47] projects that dependence on the means-tested Pensions Credit should fall from the current one-third of all pensioner households to around 3% by 2060 as a result of the more generous new State pension (nSP).nSP should improve incomes for three-quarters of pensioners retiring over the next fifteen years, with the impact remaining positive for half of those retiring later than 2050.At the same time, the end of contracting out and the introduction of The Labour administration that came in power in the late 1990s, on the other hand, took a diametrically opposite view and conducted considerable reforms, which are amply described in Hills and Stewart [44].Besides introducing a more generous system of means-tested benefits, known as Pension Credit and a more progressive State Second pension (S2P), it set up an independent Pensions Commission widely considered as one of the major policy successes of the UK government (see Rutter, Marshall and Sims [45]).This Commission advocated the joining up of the flat-rate and earnings-related state pensions into one more generous benefit, the reintroduction of earnings indexation of state pensions and the auto-enrolment of workers into workplace pensions.The latter included the UK government setting up a low-cost workplace pension scheme so that not all employers had to create their own scheme, and also to act implicitly as a competitor to private schemes.The impact of the Labour administration's pension policies has been very pronounced, such that the pensioner poverty rate nearly halved and is now at par with that in Germany.
The subsequent Coalition and Conservative administrations continued these policies, in contrast to changes effected in other social welfare policies, such as working-age benefits (more details can be found in Lupton, Hills, Stewart, Burchardt and Vizard [46]).The Coalition administration even enhanced further the indexation of state pensions to include a triple lock that ensures a minimum increase of 2.5% even if both earnings growth and inflation are below this rate.The Department for Work and Pensions [47] projects that dependence on the means-tested Pensions Credit should fall from the current one-third of all pensioner households to around 3% by 2060 as a result of the more generous new State pension (nSP).nSP should improve incomes for three-quarters of pensioners retiring over the next fifteen years, with the impact remaining positive for half of those retiring later than 2050.At the same time, the end of contracting out and the introduction of auto-enrolment appear to have given a new lease of life to the United Kingdom's occupational pension system.The Department for Work and Pensions [48] indicates that while between 2006 and 2012 participation amongst private sector employees had fallen from 51 to 42%, by 2016 it had risen to 73%.
The United Kingdom's reform experience over the last fifty years shows a clear cyclical pattern of reforms, with an initial focus on system goals turning into an emphasis on system constraints that was maintained until the achievement of system goals substantially deteriorated.Then, despite there being a pronounced financial and economic crisis, successive administrations across the political spectrum maintained the course of reversing the direction of reforms and returning to a focus on the achievement of system objectives.It is particularly noteworthy that policy continuity in other areas of welfare policy is not at all evident, and in fact, the crisis resulted in significant austerity measures in the United Kingdom for non-pensioners.Eurostat data indicate the proportion of the UK's working age population in material deprivation rose from under 4% in 2008 to close to 9% in 2013 and was still at 6% in 2016, while it halved among the pensioner population to close to 1%.
The narrative of a secular trend in pension reform-halted temporarily but then resumed once the fiscal situation requires it-does not seem to apply in the United Kingdom's case.The story seems to be that large-scale reductions in state pension generosity and an assumption that this will be made up by private saving resulted in considerable policy reversals despite a major economic and fiscal crisis.The next section will try to explain the source of this cyclical pattern in pension reform.

Optimizing the Trade-Off Between System Goals and Constraints to Make Reforms Sustainable
The case of the United Kingdom shows that while economic and fiscal conditions, together with political considerations, are undoubtedly important in driving the pension reform process, there is a more underlying driver.It appears that if policymakers carry out reforms that undermine the trade-off between the system's goals and constraints, then these reforms do not tend to be long-lasting.
Before delving more in this hypothesis, it is pertinent to present another case of cycles in pension reform.In the 1980s, Chile was the first country to adopt mandatory privately managed personal pension accounts.While Chilean officials and economists in international institutions (see Organization for Economic Co-operation and Development [49] and Iglesias-Palau [50]) criticized some features of the system, these were mainly of an operational nature and the system was seen as a model for other countries.Yet the failing adequacy of the system resulted in considerable pressures for reform.While maintaining the broad structure of the system, a number of important changes were made in 2008 (described in Mesa-Lago and Bertanou [51]), such as the introduction of a minimum pension regardless of contribution history, and disability and survivors' coverage together with credits for mothers.The number of beneficiaries doubled while participation in the system rose.These reforms, though introduced by a Socialist president, have been retained and even expanded on by her successor, who incidentally is the brother of the architect of the 1980s privatization reform in Chile.The decision to retain these reforms by a Conservative president comes despite the fact that economists from international institutions (see Santoro [52]) have argued against them on the grounds that they are expensive and will lower economic growth, and also despite the fact that the country's international rating has been downgraded because of high deficits.
Howse [34] argues that even if one accepts the argument that current levels of pension spending are high enough and that one cannot respond to the ageing transition by raising taxation above a certain level, this does not mean that the task of policymakers is simply to ensure that pension spending does not rise.For reforms to be sustainable, they need to ensure that people continue to have what they consider to be an adequate income in retirement.Simply reducing the cost of providing retirement income by moving part of it off-budget, as was done in many Eastern European and Latin American countries, has proven to be unsustainable.Governments have had to step back in to guarantee provision for low-income workers that were finding it difficult to save into private schemes.This raises questions on how much policymakers can rely on non-mandatory funded pensions to provide adequate pensioner poverty alleviation.In cases where mandatory provision is replaced by nonmandatory schemes, it is likely that those most in need will end up with little provision, requiring the introduction of other anti-poverty measures.
Policymakers need to consider the feedback effects from cuts in future pension spending.Unless individuals accept these cuts and lower their consumption during retirement or accommodate them by raising their non-pension income, the main effect of reforms could be to create pressures on other areas of government spending, resulting in no real improvement in the fiscal balance.A classic case from the United Kingdom is that while spending on pensions was reduced, government ended up introducing a plethora of other benefits for the elderly, such as subsidies for energy, transport, and medical expenses.
Holzmann and Hinz [1] is an example of the growing consensus of the need to adopt a multifaceted approach to assess pension reform, based on the realization that the underlying needs currently achieved by state pension systems cannot be simply swept away through reforms.The revised World Bank position on pension reform, reflecting the recommendations of the Independent Evaluation Group [27], is that pension systems should provide adequate, affordable, sustainable, and robust retirement income in line with conditions in each individual country.In particular, benefits need to be provided to most of the population, in a way that is sufficient to prevent old-age poverty in addition to providing a reliable means to smooth lifetime consumption for the vast majority of the population.The Social Protection Committee, an advisory policy committee for Employment and Social Affairs Ministers of European Union Member States, has also adopted a broad concept of sustainability whereby inadequate pensions are seen as a source of unsustainability, emphasizing that if the social impact of reforms is not politically acceptable, reforms would be reversed.This concept of balancing the trade-off between system goals and constraints, through which the aims of the system for current beneficiaries continue to be achieved without putting excessive pressure on future generations of workers or reducing too much their future benefits underpins the approach taken in European Commission [9].
Figure 2 depicts the interdependence between the goals and constraints faced by pension systems.On the left side one has the twin goals of the pension system.The need to ensure some degree of income smoothing can be politically important.If a system is not seen as beneficial by the electoral majority, political pressures for its reform can be difficult to resist.Similarly, if a system is not seen as effective in alleviating poverty, the political pressures that led to the setting-up of transfers to elderly people during the nineteenth century might re-emerge.The cases of the United Kingdom and Chile show that this happens even when the affected numbers are still relatively small.The likelihood of it continuing to happen in coming decades is higher as a larger share of eligible voters will be near or above pension age.
The right side of Figure 2 shows the constraints of the pension system-namely the need for it to be intergenerationally fair.A pension system should not be adequate simply in terms of the poverty alleviation and consumption smoothing it provides to current pensioners, but rather it needs to provide a fair deal to different generations.The standard conception of sustainability focuses on the need to prevent having to impose ever-increasing contribution rates to finance pension transfers.However, another source of political pressures, which is often ignored, is the desire of adjacent generations of pensioners to enjoy similar living standards in retirement.If a generation of soon-to-be pensioners believes its preceding generation had much larger pension transfers, it might pressure governments to reverse reforms that have lowered its benefits and instead push cuts to subsequent generations.electoral majority, political pressures for its reform can be difficult to resist.Similarly, if a system is not seen as effective in alleviating poverty, the political pressures that led to the setting-up of transfers to elderly people during the nineteenth century might re-emerge.The cases of the United Kingdom and Chile show that this happens even when the affected numbers are still relatively small.The likelihood of it continuing to happen in coming decades is higher as a larger share of eligible voters will be near or above pension age.The interaction between pension system goals system constraints.

Discussion
For many, the concept of pension reform conjures an inevitable process of retrenchment, either in terms of reduced benefits or rising pension ages.The only issue is its pace, seen as dependent on the current or projected state of the fiscal situation and on the strength of the political forces trying to maintain the status quo.This article has argued instead that in cases where significant changes to the pension system were made, and these resulted in much reduced achievement of system goals, the reforms tended to be reversed or else countervailing measures taken to generate better outcomes for pensioners.This suggests that the sustainability of pension reforms requires that pension adequacy and fiscal affordability be treated as two sides of the same coin.While earlier papers, such as Holzmann and Hinz [1], have argued this simply from a theoretical perspective, this article has sought to provide a more empirical approach focusing on evidence from the United Kingdom and Chile, two of the countries where the focus on fiscal considerations has been reversed in recent reforms.The article argues that in future there could be other similar cases, particularly when the impact of reforms on pension adequacy starts to be more strongly felt.
Eckardt [53] argues that fiscal considerations predominate because there have been few attempts to develop prospective income indicators to evaluate the effects of reforms.Since then, this lack has been gradually addressed (Zaidi and Grech [54]; Buslei et al. [55]; European Commission [9]).Policymakers are starting to give due consideration to pension adequacy.Even in cases where benefit generosity has been reduced considerably, minimum pensions have been strengthened.The prospective effect of reforms on low-income households and on women is becoming an integral part of impact assessments.
That said, there continues to be a need to undertake a more holistic approach to pension reform.For instance, simply raising pension ages while doing little to help older workers remain in employment, such as providing assistance to boost their productivity, will not necessarily reduce the burden on the public purse.Similarly, the Chilean case shows that providing incentives for people to save in private pensions but doing little to ensure they provide good value for money, is a policy with a short sell-by date.To limit the recurring cycle of pension reform, and make reforms sustainable, policymakers need to ensure that the broader policy environment is conducive to a better achievement of the pension system's goals within its constraints.

Figure 1 .
Figure 1.The cycle of pension reforms in the UK.

Figure 1 .
Figure 1.The cycle of pension reforms in the UK.

Figure 2 .
Figure 2.The interaction between pension system goals and system constraints.

Figure 2 .
Figure 2.The interaction between pension system goals system constraints.