The future of pensions: reforms and their consequences introduction R1

Transcription

1 The future of pensions: reforms and their consequences introduction R1 THE FUTURE OF PENSIONS: REFORMS AND THEIR CONSEQUENCES INTRODUCTION Alexander M. Danzer,* Richard Disney,** Peter Dolton*** and Chiara Rosazza Bondibene**** Origins of pension systems Relatively generous provision of incomes for old people is among the main achievements of modern westernised economies. Historically, the state entered and expanded the market for pension provision in order to eliminate old-age poverty and to overcome diverse types of market failures. However, since pension budgets have been increasingly challenged by demographic and financial developments, governments around the world have been reforming their pension systems. This special issue sheds light on the origin, design and potential consequences of pension reform with a special focus on the UK and Germany. Before turning to the structure and reforms of specific pension systems, we first discuss reasons for why governments should engage in pension provision in the first place. As the basis of this analysis we utilise the standard framework that identifies five rationales for state intervention (Diamond, 1977): market failure, paternalism, redistribution, revenueraising and administrative efficiency. While these rationales for the public provision of pensions have supported the historical development of public pension systems, the challenges of adequate coverage and fiscal sustainability have contributed to the emergence of highly complex pension systems. Most countries nowadays benchmark their pension systems against a standard three-pillar system (as also proposed by the World Bank for developing and emerging economies): a basic state-provided pension pillar with the goal of poverty reduction, a second (often funded) pillar comprising employer co-sponsored pensions and a third pension pillar comprising voluntary privately funded pension plans promoting a stronger relation between earnings and pensions (as the rich can contribute more to their private accounts). Market failure Market failure may arise from incomplete information both prior to, and as a consequence of, insurance contract implementation. Adverse selection in its various forms in annuity markets is often an obvious candidate requiring public intervention. While there is some evidence of adverse selection in UK annuity markets (see the literature cited in the paper by Disney in this Review), adverse selection is generally seen as a rationale for compulsion in insurance rather than for public provision per se. Paternalism In contrast, the paternalism argument suggests a role for state intervention on the grounds that incomplete information (and irrationality) induce households to make incorrect decisions either manifestly timeinconsistent or inconsistent with the preferences of the social planner. Of course, complexity of pension provision may be a cause of sub-optimal household decision-making as well as a consequence of it. The commonest assertion in this context is that households typically undersave for retirement relative to an optimal plan of lifetime saving. This argument has often been adduced in official documents by applying rules of thumb to optimal saving (for example, an arbitrary benchmark as to what should be *KU Eichstätt-Ingolstadt, IZA Bonn, CESifo Munich; **University of Sussex, University College, London, Institute for Fiscal Studies; ***University of Sussex and CEP, LSE; ****National Institute of Economic and Social Research and Centre for Macroeconomics. c.rosazza_bondibene@niesr.ac.uk.

2 R2 National Institute Economic Review No. 237 August 2016 the replacement ratio of the average pension to average lifetime earnings and a relatively narrow definition of what constitutes retirement saving an example of this approach being the Pensions Commission, 2004). However, estimates of the optimality of saving based on a more explicit model of life-cycle saving with consumption smoothing tend to cast some doubt on the assertion, especially when non-pension assets are included in the definition of saving (see Banks et al., 2005; Crawford and O Dea, 2014). In addition, however, the paternalist case for intervention need not necessarily extend to state provision as such. Instead, state provision may distort the decisions of savers (Diamond, 1977). Disney (2006) finds some support for the crowding-out of private by public saving by hypothesising it will be greater in countries and time periods where the design of the social security system more closely mimics a private retirement saving instrument. Alternative microeconometric tests of whether individuals respond to changes in retirement saving incentives in a manner consistent with consumer rationality have delivered inconclusive results. Most evidence in the UK suggests that individuals do respond to retirement saving incentives in a rational manner, once those incentives are correctly specified, but these behavioural responses are fairly modest and individuals are subject to a high degree of inertia. This inertia may suggest a greater role for pension education and for novel types of intervention to persuade individuals to save. Redistribution The argument concerning redistribution towards pensioners is often confused by a conflation of various definitions of redistribution. Arguably, simply comparing the average income of pensioners at a point in time with the average income of workers is a misleading perspective and such comparisons should instead be presented over a lifetime perspective. Hence one measure of redistribution is intergenerational comparing, over the life cycle, the receipt of benefits relative to taxes and contributions paid (perhaps formally computed as an average internal rate of return ) of successive date-of-birth cohorts. Almost all evidence for most countries, including the UK, is that such returns were highest for those brought into the public pension programmes in their inception stages (typically in the 1950s) and again among the early baby-boomers (Disney, 2004; Disney and Emmerson, 2005) compared to the current working generations, given demographic ageing and other cutbacks. In private pension plans, this decline in returns to later cohorts may be replicated given low annuity rates, falling support ratios of contributors to pensioners and the sluggishness of capital markets. In Germany, attempts were made to balance the financial burden for increasing longevity and shrinking cohorts of pension system contributors between the younger and older generation. The introduction of private old age provision (Riester Reform 2001) was meant to relieve the pressure from young cohorts who face declining replacement and higher contribution rates (through the so-called sustainability factor introduced in 2004). At the same time, the greater statistical longevity of younger workers has been turned into longer working lives (increasing the normal pension age gradually from 65 to 67, legally implemented in 2007). Hence, unlike the UK, Germany has only fifteen years of experience with a legally implemented comprehensive three-pillar system; before that pension entitlements were almost entirely accumulated in a monolithic social security PAYG insurance system (Börsch-Supan and Wilke, 2005). Still, the first pillar remains dominant as the cohort born in 1970 will generate around 70 per cent of their pension income from the social security system even if everyone contributed to the private pension pillar with publicly promoted amounts (Börsch-Supan et al., 2008). In fact, the take-up of pension contracts from the third pillar remained below expectations. This not only implies a greater future weight for the first pillar of the pension system in Germany, but also lower predicted pension levels among the younger generation. In general, the financial position of the German pension system depends, in great part, more on the performance of the labour market than the UK system (Bonin, 2009). The question of whether pension programmes are more or less redistributive within a generation across the lifetime income distribution that is, intra-generational redistribution is more complex. A progressive pension programme (for example, one with a flat element to benefits combined with a progressive schedule of contributions or tax payments) is certainly more redistributive than a programme based on earnings replacement (that is, one with pensions positively related to lifetime earnings) but the overall progressivity of the programme will depend on this design aspect versus the mortality divide across the income distribution. This is illustrated by Creedy et al. (1993) for the UK programme, who argued that the progressivity of the pension formula in the public programme at that time was more or less cancelled out by differential occupational mortality rates across the income distribution. Intra-generational redistribution

3 The future of pensions: reforms and their consequences introduction R3 is relatively low in Germany as benefits in the social pension system (first pillar) are earnings related. As a consequence, a major challenge for the German pension system is the comparatively low replacement rate for the low educated as they most often do not benefit from the second or third pillar of the new system with poor entitlements in the first pillar (Bonin, 2009). Revenue-raising The revenue-raising rationale for state intervention relies upon the premise that individuals are more willing to pay what they believe to be contributions to their own future well-being than taxes or payments in general, however illusory this belief may be. This argument is most applicable to public programmes although it also applies in defined benefit (DB) pension plans, where contributions and benefits are, at best, loosely related. However, the idea of revenue-raising has most salience in relation to public social insurance programmes which contain a contributory principle, such as the core UK National Insurance programme and the German public programme. Such programmes are generally effectively tax-and-transfer (PAYG), with no fund accumulated, so that current individual contributions are not saved in an explicit accumulated fund. Hence payouts to future generations of pensioners may or may not reflect the contributions that they made in their working lives. In any event, within-generation payouts of pensions bear little relation to contributions on an individual basis, as just described. Given the illusion at the heart of the contributory principle, many advocates of tax and pension reform in the UK have called for the abolition of the distinction between National Insurance contributions and income taxes as they are causing the unnecessary duplication of functions and complexity in the personal tax structure (see, e.g., Dilnot et al., 1984). But the continued co-existence of the National Insurance system with income tax may in part suggest that successive governments believe that a separate contributory principle is an important device to raise money to pay current pensions; hence, it is unlikely to be abolished. Bringing the rules of National Insurance more closely into alignment with the remainder of the personal income tax structure would be a useful compromise (Mirrlees, 2011). Administrative cost The relative administrative costs of a comprehensive public pension system to a more personalised private provision may seem a rather mundane rationale for preferring one form of pension programme to another, but it certainly exercised for example Sir William Beveridge in his famous report Social Insurance and Allied Services (1942) and the Thatcher administration in its early years. Clearly a larger, or nationally-provided, system could be provided at lower administrative cost than individually-purchased retirement saving accounts. This cost reduction should arise from economies of scale and other efficiency gains, especially from eliminating the scope for adverse selection. However, some of the higher costs of private provision may arise from the greater choice of investment options available in an individually-tailored plan rather than the one size fits all approach which is inherent in a national programme. Challenges and reforms of pension systems The first two articles of this special issue present a comprehensive overview of the UK and German pension systems; while the UK system broadly corresponds to the three-pillar logic with a relatively strong private provision, the German system predominantly relies on the first pillar: a state-provided pay-as-you-go (PAYG) insurance system with (too) little private provision. The historical evolution of these pension systems can be understood against the social necessities of the 20th century and the limitations of private old-age provision. In his overview of the UK pension system Disney presents the UK programme as having had four stages of post-1945 development with no certainty that the reform process is yet complete. Werding gives a comprehensive overview of the German pension system, which has undergone reforms more recently. Both overview chapters illustrate that today s political complexity in pension reform partly stems from the coexistence of several pension pillars which have developed independently but are also intertwined in an interdependent scheme of old-age provision. As a consequence, politicians have remained reluctant to pursue comprehensive reforms and instead have redesigned single pillars. Despite the distinctively different set-up of their pension systems, the UK and Germany face similar demographic challenges and financial conditions: as the population ages, an increasing number of retirees draw their pensions for longer, while contributing cohorts are shrinking. This makes PAYG systems harder to sustain in the long run without politically unacceptable tax rises. This has led both countries to reform unfunded systems as well as to promote funded pension systems. Yet, funded systems have been facing financial difficulties in times of low nominal returns on safe assets. In addition, the slow growth of annuity

4 R4 National Institute Economic Review No. 237 August 2016 markets coupled with low annuity rates as a result of demographic ageing have cast doubt on the vision that a transition to private pension provision can make oldage provision more sustainable. Such diverse challenges have profound consequences on the distribution of pension wealth within countries as different pension systems often coexist even within the same pension pillar. For instance, until recently the second pillar of the UK system comprised a mandatory earnings-related pension scheme for employees. Employees could contract-out of this system if they (together with their employers) made equivalent contributions to a funded defined benefit (DB) or defined contribution (DC) pension scheme. Matters are different for public service employees whose DB pension schemes are mostly unfunded (with the exception of the Local Government Pension Scheme). In effect, employees of a specific occupation could have widely different pension entitlements depending on whether they were contracted-in or out, their sector of employment and their choice of pension plan. In the German system, two independent pension schemes coexist. While private sector employees finance the PAYG branch (first pillar) of the pension system with social security contributions, public employees (Beamte) are not contributing while receiving more generous pensions, either from a Länder-specific pension fund or from tax revenues; they pay, in exchange, higher taxes on their pensions and have to co-fund their health care plans during retirement. While the goal of redistribution was among the fundamentals of pension provision (see first section), today s pension schemes face problems of both intergenerational and intragenerational inadequacy. While the former challenge can be met by increasing contribution rates, postponing retirement ages and/or cutting benefit levels, the latter implies moving towards flatter benefits levels and the strengthening of provision for low-income earners as anti-poverty insurance. Addressing some of these concerns for Germany, Vogt and Althammer simulate the effects of different policy reforms on the financial sustainability of the pension system. The UK has become an especially active study field for pension reform. This special issue contains a number of papers reviewing the most recent reforms of the UK system. One major reform has been to reduce the relative generosity, and hence costliness, of public sector pensions, while at the same time reducing the intra-generational inequality of outcomes. Cribb and Emmerson analyse the impact of the Hutton reform on the relative pensions of representative public vs. private-sector workers. Major redesign elements were the transition from a final salary to a career average pension computation and the uprating of benefits using the CPI rather than the RPI. Finding a decline in the wealth of (especially high paying) public sector pension plans has important implications for the attraction and retention of workers in the public sectors. Danzer, Dolton and Rosazza Bondibene account for the substantial heterogeneity in pension plan reforms across different public sector occupations following the Hutton report. They illustrate how differently members of various public sector occupations were affected by an intentionally homogenous public sector pension reform. In addition to public sector DB pensions, the DC pension system has come under reform in the UK: Cannon, Tonks and Yuille analyse the effect of abolishing mandatory annuitisation of DC pension pots in the UK in 2015 on the market for annuities. They suggest that there will be major repercussions in the market as a consequence of this surprising policy change. Weale and van de Ven discuss possible explanations for the policy change, which was surprising in the sense that it may now encourage individuals to engage in myopic savings behaviour. The picture emerging from the collection of papers in this special issue suggests the importance of carefully designed pension reforms that account for the potential distortions that such changes might create. These distortions might take the form of counterproductive or unanticipated forms of redistribution between groups of pensioners as well as workers within different generations. Moreover, the redesign of the pension system should perhaps be more comprehensive and comprehensible, given that current reform attempts remain largely fragmented and deal with problems of sustainability or affordability within single pillars. A holistic approach to pension reform will also need to take into account potential switches between the private and public sector occupations, the differences in workers education and in their financial literacy.

5 The future of pensions: reforms and their consequences introduction R5 REFERENCES Banks, J., Emmerson, C., Oldfield, Z. and Tetlow, G. (2005), Prepared for Retirement? The Adequacy and Distribution of Retirement Resources in England, Institute for Fiscal Studies Research Report No 67, London. Beveridge, W. (1942), Social Insurance and Allied Services, 847, London: HMSO. Bonin, H. (2009), 15 Years of Pension Reform in Germany: Old Successes and New Threats, IZA Policy Paper No. 11. Börsch-Supan, A., Bucher-Koenen, T., Reil-Held, A. and Wilke, C.B. (2008), Zum künftigen Stellenwert der ersten Säule im Gesamtsystem der Alterssicherung, DRV Schriften, 80, pp Börsch-Supan, A. and Wilke, C.B. (2005), Shifting perspectives: German pension reform, Intereconomics, 40 (5, September/ October), pp Crawford, R. and O Dea, C. (2014), Cash and pensions: have the elderly in England saved optimally for retirement?, Institute for Fiscal Studies Working Paper W14/22, London. Creedy, J., Disney, R. and Whitehouse, E. (1993), The earningsrelated state pension, indexation and lifetime redistribution, Review of Income and Wealth, 40, (September), pp Diamond, P. (1977), A framework for social security analysis, Journal of Public Economics, 8, pp Dilnot, A., Kay, J. and Morris, N. (1984), The Reform of Social Security, Oxford: Oxford University Press. Disney, R. (2004), Are contributions to public pension programmes a tax on employment?, Economic Policy, 39, July, pp (2006), Household saving rates and the design of public pension programmes, National Institute Economic Review, 198, October, pp Disney, R. and Emmerson, C. (2005), Public pension reform in the United Kingdom: what effect on the financial well being of current and future pensioners?, Fiscal Studies, 26 (March), pp Mirrlees, J. and IFS (eds) (2011), Tax by Design. The Mirrlees Review, Oxford: Oxford University Press. Pensions Commission, The (2004), Pensions: Challenges and Choices, The First Report of the Pensions Commission, London: the Stationery Office.