RAIN Audrey

ACKNOWLEDGEMENTSThis work was done in the context of a research partnership between theInstitutdes politiques publiquesand theDirection de l’Animation de la Recherche, des Etudeset des Statistiques(Dares) at the French Ministry of Labor. We thank the Dares forits financial support that made it possible to realise this study.Our partners at the Dares have also provided several suggestions that dramat-ically improved the quality of the report. We would like to thank in particular fortheir useful feedback Cécile Ballini, Mathilde Gaini and Philippe Zamora. We alsothank them warmly for their patience and flexibility throughout the project.We then thank warmly the IPP team for constant support with the project. Weare particularly grateful to Julien Grenet as he suggested the central idea that isinvestigated in Chapter 4 of this report, and provided insightful comments through-out the project. Marion Monnet also deserves to be thanked for participating in themanagement of the project at its initial stage, and helping throughout.This report was written after the organization by the Dares and the IPP of aworking group on skills and skill mismatch. We would like to thank all participantsand their institutions for their feedback when we presented preliminary versionsof our work to the group. We also thank the OECD which allowed us to access aslightly enriched version of the PIAAC data, which was very useful to undertake theanalyses presented in the report.

A consequence of the French pensions reform whose aim is to establish a universal pension system having defined yield and operating on a points basis will be to reinforce the contributory nature of the formula for calculating pensions. Whereas in the current system the contributory core has counter-redistributive effects – increasing the pensions inequality relative to the salaries inequality – the new system would become neutral and the reform would thus lead to a reduction in pension inequalities. The reason for this counter-intuitive effect – i.e. the effect whereby making the system more contributory reduces inequalities – is to be found in the corrections made by implicit mechanisms in the current system, such as the rules of taking the 25 best years or of revaluating the salaries included in the pensions calculation in line with inflation. Abolishing the rule of number of years of contributions in the pensions scale would also reinforce this effect by being more beneficial to individuals who have had low mean salaries. In this policy brief, we show these effects based on simulations conducted on the population of employees in the French private sector. In addition to individuals on low salaries, women would also benefit signicantly from this change in the calculation formula.

An important feature in the debate on French pensions reform is whether or not it is necessary to keep a reference retirement age in the new system. This brief aims to contribute to the debate by clarifying certain ambiguities about the concept of retirement age and by discussing the potential implications of implementing a points system. We stress the difference between the impact of reference ages in the current system – those ages changing the pension scale of the system – and implementing reference points in the new system, such points playing a useful part in informing the future pensioners. Recent economic literature has highlighted the part played by reference points in pension scales, beyond providing purely financial incentives. This would argue in favour of the new system keeping a target to which the future pensioners can refer. Rather than a single pivotal age for everyone, this brief advocates introducing a reference norm that is defined by obtaining a target replacement rate, e.g. 75 % of the last salary before retirement. Such a reference would lead to dening an individual full-pension age, adapted to each career. This would also be a return to the initial goal of a pension system, namely to maintain standard of living on retirement. Such an age reference could also be accompanied by new services for helping future pensioners prepare their retirement choices better.

An important element of the pension reform debate is whether or not the new system should maintain a reference age. This note aims to contribute to this debate by clarifying some of the ambiguities around the concept of retirement age and by discussing the potential implications of implementing a points-based system. We emphasize the difference between the impact of reference ages in the current system - which modify the scale of the pension system - and the introduction of benchmarks in the new system, which can play a useful role in informing the insured. Recent economic literature has highlighted the role played by pension benchmarks, beyond purely financial incentives. This argues for maintaining a target in the new system to which the insured can refer. Rather than a single pivot age for all, this note argues for introducing a reference standard that is defined by the achievement of a target replacement rate, for example 75% of the last salary. Such a benchmark would lead to the definition of an individual full rate age, adapted to each career. It would also restore the initial objective of a pension system, which is to ensure that the standard of living is maintained in retirement. Such an age reference could also be accompanied by new services for insured persons to better prepare their retirement choices.

This study aims to simulate a pension system based on points, with a view to the reform in preparation aiming at the introduction of a universal points-based system. The objective of this report is to focus on the definition of the rules for steering the system. In particular, what are the golden rules of the system that should be the subject of strong commitments by the legislator to guarantee pension rights? To what extent can part of the steering be subject to more flexible decisions, as the system evolves? The methodology of this work is based on a theoretical analysis and the simulation of demographic and economic shocks in order to study the consequences for the sustainability of the system and the level of pensions of different steering rules. It involves subjecting the pension system to stress tests, similar to the banking stress tests simulated by central banks.

A points system, operating at defined yield, makes it possible to rethink how pension systems are managed. Instead of having to make repeated ad hoc changes to the parameters of the system, it is possible to define change rules that other guarantees to future pensioners, as regards not only their entitlements but also the long-term sustainability of the system. In this brief, and based on simulations of a variety of shocks to the pension system, we study what management rules deserve to be chosen. Two rules absolutely must be selected: firstly the growth in the value of the pension point should match the growth in salaries. and secondly converting the points into pension should take into account the life expectancy of each generation (cohort). A third rule that is important for the long term, is the relationship between the rules for index-linking claimed pensions and the amounts of the pensions when they start being claimed. This rule should serve as a guide to managers so that they can steer the system towards an equilibrium that is not based on too low an index-linking of the pensions. Such management implies high institutional autonomy for the system, whereby the managers need to be accountable for the finnancial equilibrium and for the risks to pension revaluation.

The pension reform aiming to introduce a universal pension system with a defined return in points will have the consequence of reinforcing the contributory character of the pension calculation formula. While in the current system the contributory core has anti-redistributive effects - increasing pension inequalities relative to wage inequalities - the new system would become neutral and the reform would thus lead to a reduction in pension inequalities. The reason for this unintuitive effect - an increase in contribution reducing inequalities - is the correction of mechanisms implicit in the current system, such as the rule of the best 25 years or the revaluation by inflation of the wages included in the calculation of pensions. The abolition of the insurance duration rule in the scale also reinforces this effect by benefiting relatively more individuals with low average wages. In this note, we highlight these effects based on simulations carried out on the population of private sector employees. In addition to low-wage individuals, women would largely benefit from this change in the calculation formula.

A points-based system, operating with a defined return, makes it possible to rethink the management of the pension system. Instead of having to repeatedly modify the system's parameters on an ad hoc basis, it is possible to define rules of evolution that offer guarantees to the insured, both on their rights and on the long-term sustainability of the system. Based on simulations of a variety of shocks to the pension system, this note examines which steering rules are worth retaining. Two rules must absolutely be retained: first, the value of the pension point must follow wage growth, and second, the conversion into a pension must take into account the life expectancy of each generation. A third rule, important for the long term, is the relationship between the rules for indexing liquidated pensions and the amount at liquidation. This rule must serve as a guide for managers in order to steer the system towards a balance that is not based on too little indexation of pensions. Such steering implies a strong institutional autonomy of the system, where managers must be responsible for the financial balance, as well as for the risks on the revaluation of pensions.

The Institute of Public Policy (IPP) is developed in the framework of a scientific partnership between PSE and CREST. The IPP aims to promote the quantitative analysis and evaluation of public policies by using the most recent methods of economic research. Abstract The pension reform announced by Mr. Emmanuel Macron during the presidential campaign is based on a certain number of general principles, often summarized by the commitment "each euro contributed gives the same rights". As the details of the reform are not yet known, this note aims to contribute to the public debate by presenting the fundamental principles of a well-designed pay-as-you-go system, regardless of the target system chosen, and to highlight the points of arbitration to be decided by the democratic debate. We thus highlight the advantages of increasing pension entitlements in line with wage growth, and of establishing transparent rules for the evolution of settlement rates according to the country's demographic conditions. The trade-offs to be made are nevertheless important: what overall contribution rate to choose? what convergence between the contribution rates of the different schemes? how fast to transition to the new system? how to improve solidarity mechanisms? or what governance to put in place? - The debate on the choice of the ideal pension system cannot be reduced to the opposition between annuity, notional accounts or points-based systems, but should be based on general operating principles. - The revaluation of rights acquired during a career according to wage growth is the only option that guarantees that each euro contributed gives the same right regardless of when it was contributed. - For a given contribution rate, the financial equilibrium of the system also requires adjusting the rate at which pensions are paid out to demographic conditions, in particular the average life expectancy per generation. - Solidarity mechanisms can be designed to reproduce the redistribution of the current system, or even to improve it.

The pension reform announced by Mr. Emmanuel Macron during the presidential campaign is based on a number of general principles, often summarized by the pledge "each euro contributed gives the same rights." As the details of the reform are not yet known, this note aims to contribute to the public debate by presenting the fundamental principles of a well-designed pay-as-you-go system, regardless of the target system chosen, and to highlight the points of arbitration to be decided by the democratic debate. We thus highlight the advantages of increasing pension entitlements in line with wage growth, and of establishing transparent rules for the evolution of settlement rates according to the country's demographic conditions. The choices to be made are nevertheless important: what overall contribution rate to choose? what convergence between the contribution rates of the different schemes? how fast to transition to the new system? how to improve the solidarity mechanisms? or what governance to put in place?

The work presented in this thesis focuses on the individual returns to education and training, and seeks to identify more effective ways in which public intervention can increase these returns. The first two chapters of this work explore how returns to individual investments in education can be maximized by improving the efficiency of education and vocational training systems. The final study examines the links between human capital investment and the legal or economic environment that frames it. The first article of this thesis focuses on the effect of enrollment in a French private school in CP and CE1 on academic performance in CE2. The second chapter aims to measure the effectiveness of certification training for French job seekers on their return to employment. The last study shows the link between the flexibilization of the English labor market and access to training for employees. The analyses carried out are based on microeconometric methods which aim to identify the causal effect of the public policies studied. We thus use the instrumental variables method and the difference-in-differences method. This thesis also relies on the estimation of duration models, using the timing-of-events method or estimating a bivariate competing risks model.

This study approaches the comparison between public and private schools from two perspectives. The first is to better understand the choice of sector at the beginning of elementary school (CP and CE1). This choice differs according to the family environment, the proximity of a private school to the home, and the level of the student at the beginning of CP. The second aims to judge the effectiveness of the private sector in helping students succeed. A direct comparison of results between the two sectors is not relevant, as their students do not have the same characteristics. To account for these differences, we use information collected from the panel of students who entered the first grade in 1997. The difference in distance to the parents' home between the nearest private and public schools is used as an instrumental variable. Overall, we find no significant effect of private school attendance in CP and CE1 on the results at entry into CE2, in mathematics as well as in French, nor on the probability of repeating the first and second grade.