Leader: Gustavo De Santis (UNIFI); Other collaborator(s):
As survival conditions change, the question arises of how properly adjust the retirement age. The question lies at the intersection of demography (when do we become old?), medicine (how long can we be productive, and for what kind of occupations?), and welfare policies (how to maximise social wellbeing?), and it has very relevant practical implications. The solution adopted in Italy (retirement age increases with life expectancy at age 65) appears to be theoretically weak. Are improvements possible? Beyond this, considering that longer survival is not the sole cause of ageing, and not the sole cause of concern about the viability of the pension system, can improvements be introduced also in other respects?
Brief description of the activities and of the intermediate results: A colleague and I have recently presented evidence suggesting that the standard demographic interpretation of ageing being caused by low fertility may be biased. While low fertility does matter on relatively short-term changes, it is survival conditions that matter more in the long run. And these can be simplified: indeed, the current, time-varying life table, and in particular the age structure of the embedded stationary population, can be used as a guideline for understanding the where the current age structure is heading. This finding may appear to be of limited interest, because the “force of attraction” that the stationary age structure exerts on the current age structure is small, and the time required for its effects to emerge may be long (several decades). However, it become crucial when applied to pension systems, because in that case the “natural” time horizon is (or at least should be) very long.
Indeed, this finding goes hand in hand with other ideas aimed at creating a viable pension system, i.e. capable of resisting unchanged to future changing circumstances, both in demographic and in economic terms. This system, named IPAYGO (Improved Pay-As-You-Go), is currently being elaborated and will form the object of at least two future publications. One is about to become public, because a scientific paper has been accepted for publication in the journal “Statistics. Politics and Policy”: De Santis (2024) “Demography, economy and policy choices: the three corners of the pension conundrum”.
More details about these ideas will be given in a book on the pension systems, currently in preparation (in Italian, of approximately 100 pages, possibly to be published by Il Mulino, a prestigious publishing house in Italy). It will show how to use a relative approach to “absorb” all possible demographic and economic shocks, and therefore preserve the essential (collectively decided) characteristics of a pension system in all possible demo-economic scenarios.
Main policy, industrial and scientific implications: The policy implications of the above-described scientific findings are potentially huge. The most important is the proposal of a new pay-as-you-go (PAYGO) pension system, which proves better than all others (including the highly praised NDC - Notional Defined Contribution) in practically all possible respects. E.g. 1) it is viable in all demo-economic scenarios (thanks to a series of AAMs - Automatic Adjusting Mechanisms); 2) can be modelled parametrically, to take different shapes and fit the needs/preferences of potentially all world societies (es. in terms of length of time spent in retirement, or relative generosity towards the retirees); 3) incorporates the locally preferred degree of actuarial equity and redistribution towards the poor; 4) can be adapted to include transfers towards the young (with beneficial effects on the functioning on the system, including a small “fertility premium”); can be adapted to incorporate a period of part-time work / half pension, and thus facilitate a smooth transition to retirement. Simulation proves that it performs better that its alternative on a number of criteria (es. “cross-sectional equity”, or replacement rate; stabilization of the equilibrium contribution rate; and minimization of quasi-capital gains and losses).
The system can be presented in an easily-understandable way to the general public, under the heading “Everything is relative” - where everything means literally everything: length of life and the shares of it to be spent in the three basic conditions of young, adult and senior; current production, and equilibrium between actuarial equity and redistribution. While simple it practical application. The system relies on a very profound and complex background, both in demographic terms (e.g. with its use of a “reference” age structure), and in economic terms (e.g. with the introduction of the new, but very useful concept of adults’ net labour earning, the economic numeraire of the entire system).
This unit has worked intensively on the assigned task (connecting retirement age to survival modifications, and notably improvements) but has extended its scope, and has succeeded in designing a comprehensive pensions system that solves this (ageing due to longer survival) and several other problems. The proposed system has a number of desirable properties (e.g., a balanced budget is always guaranteed, and so are the basic, pre-defined societal goals), among which one of the most important is that it automatically adapts to all possible demographic and economic scenarios (which implies, among other things that it does not need forecasts). This construct is, understandably, very complex (albeit relatively simple, given the problem at hand, and in all cases simpler than all other existing alternatives) so, after presenting it to the scientific community (at the European Population Conference, June 2024 and in a scientific paper, De Santis 2024 - see repository), the task is now mainly that of diffusion: to stakeholder, policy makers and the general public.
Some of the basic ideas are presented in the policy brief recently prepared by this WP (see repository). Others are illustrated in two public engagement articles soon to appear in the N-IUSSP blog https://www.niussp.org/ ; others still in an article just accepted for number 4/2024 of the Journal “Il Mulino” (in Italian).
Finally, and perhaps most importantly, after several months of hard preparation, a book (in Italian) has been finalized on this topic (pension systems; how they normally work, where and why they generally fail and how they could be improved) and submitted to “Il Mulino” (perhaps the most important publishing house for these topics, in Italy). An answer about its acceptability is awaited in the next few months.
The work on the assigned task continued: How to connect retirement age to survival modifications (improvements), with a broader scope than originally foreseen. No longer just about retirement ages, but all relevant threshold ages; and not just about ages: it is a whole new pension system that has been conceived and proposed to the Italian and international community. The proposed system has a number of desirable properties (e.g., a balanced budget is always guaranteed, and so are the basic, pre-defined societal goals), among which one of the most important is that it automatically adapts to all possible demographic and economic scenarios (which implies, among other things that it does not need forecasts). These ideas have been presented in conferences and articles (see repository) and in 2025 will take the form of a book: after a few months of contacts and discussion, a contract has been signed with “Il Mulino” (perhaps the most important publishing house for these topics, in Italy), and my book (in Italian), titled “Nati con la pensione”, will be published in 2025 (probably July).
Participation in the Age-Florence opening, 19 December, 2024. Presentation of a poster (Longevità = invecchiamento? Non necessariamente...) by Gustavo De Santis.
In the first trimester of 2025, this task force worked along two main lines. The first is the finalization of the book on pensions referred to in the previous reports. The contract has been signed with “il Mulino” (probably the most important publishing house in Italy for this type of literature), and the manuscript handed in, but a series of small details and adjustments still proved necessary. After discussing the generalities about pension systems (philosophy, problems, constraints, purpose, ...) and the limits of existing theoretical models (defined benefit, defined contribution, points systems, NDC) the book presents a new proposal (IPAYGO - Improved pay-as-you-go) that seems to outperform its alternatives in all possible senses (sustainability, actuarial equity, redistribution, transparency, ...), including the technical possibility of its practical application.
The second line of work refers to the study of small scale mortality in Italy. Mortality is one of the factors that shape the age structure (and the most important one in the long run): however, assuming that the same mortality schedule applies to the entire nation, as studies of the pensions systems frequently do, may be unwarranted. Information on mortality differential at geographic level is available in Italy down to provinces (NUTS3). Nothing official exists at finer level, but, with a few colleagues, I elaborated a system to estimate it a municipality level. The results of this idea, applied to Italy in the years 2002-2018, will be presented to two forthcoming conferences: AISP’s Popdays (Cagliari, 4-6 June 2025) and IUSSP’s International Population Conference (Brisbane, Australia, 13-18 July 2025).
Coming soon