As Franco Modigliani theorized nearly 70 years ago in his Life Cycle Hypothesis, for the large majority of the population, economic security in old age depends primarily on savings from labor income earned during the working period of their life. This is so independently of the methods of pension financing (either public pay-as-you-go or private, financially funded, or any combination of the two); as well as of the Bismarckian vs. Beveridgean features of the pension formulae. What is not ‘earned’ through savings/payroll taxes and returns from capitalization is either assistance/implicit compensation – if it helps unlucky workers or represents public recognition of unpaid social activities – or a privilege, if it derives from granting higher returns to more highly paid workers (or categories with more political clout) than to lower paid ones.

At the time Modigliani wrote, the typical life cycle was centered on the husband as ‘family breadwinner’, with his spouse in the background, a supplementary income provider, whose participation in the labor market suffered many interruptions due to maternity, child rearing, care activities, and unemployment spells, more frequent for women as their jobs were (and still are) typically more precarious than those of men. In such circumstances – largely determined by social norms and dominant ‘values’ – the economic situation of women was directly dependent on their husband's pay and financial wellbeing as well as on ‘paternalistic’ rules governing both family arrangements and the social security systems. This socio-economic model – epitomized by the ‘Scottish Widows’ Fund’ established as early as 1812 to provide for widows, sisters, and other female relatives of workers in case of their premature death – was prevalent almost up to the present day. It gave little or no value to women's economic independence.

It is no surprise, then, that even today, notwithstanding profound changes in the economic structures and activities, in Organization for Economic Co-operation and Development (OECD) countries, women aged 65 or more ‘receive on average, from the pension system, combining public and private sources, 26 per cent less income than men in the same age group’. The average, as always, hides significant variance between different countries, for example from 3% in Estonia to almost 50% in Japan, two very diverse societies. These and many other stylized facts are documented by an interesting, recent OECD study on the gender pension gap, its cross-country/time variations and its causes, combining different perspectives, analytical tools and policy recommendations.Footnote ¹

Although Social Security is always a mix of (longevity) insurance/redistribution/public recognition of social rights and political concern – for example, it normally includes compulsory coverage for survivors, mostly widows – private schemes provide insurance either as a market product or as a result of informal family exchanges and acknowledged family duties (‘implicit contracts’). Given that the private sector has traditionally been less sensitive to issues of gender parity, it is also no surprise that in countries relying more heavily on private provisions, such as the USA, widows still represent a typical ‘at risk of poverty’ social group (not to mention countries outside the OECD were widows have almost no rights at all, sometimes not even to work and provide for themselves!).

The general emerging picture depicts women as working less than men (both in number of years and in hours per year), contributing less and saving less for their pensions (not necessarily as a percentage of their working income, although the OECD study refers, without adequate empirical support, to women's higher propensity to consume); retiring earlier.

Beyond these more immediate causes of the gap, the OECD study investigates social, psychological, and cultural factors, with a lot of consequent intricacies hardly allowing a proper causal analysis. Differences between genders start at the beginning of life, with emphasis on ‘masculine’ toys for boys and ‘feminine’ ones for girls; incentives toward different school subjects (more mathematics, sciences and financial education for the boys; more arts and literature for the girls); subtle stereotypes and language biases which tend to create familiarity – or, conversely, detachment – with issues such as earning money, saving, facing risk, budgeting, and the likes. All these can have important consequences both on attitudes/preferences and on choices later in life, as a rich literature (referred to in the report) confirms. Later, there are all the disadvantages/discriminations in labor market participation, career progression (the ‘glass ceiling’), and earnings, which translate in disadvantages in pension accruals and entitlements.

Subtler disparities are also generated by pension rules, such as eligibility conditions, retirement ages, and the specific formulae to calculate benefits. Sometimes these explicitly and paternalistically favor women, with the aim of providing an ex post compensation for ex ante discrimination, as in the case of earlier retirement ages, a ‘generosity’ that backfires on the pension level both directly – through the lower age – and indirectly – through less efforts (and often anti-female prejudice) in career advancement resulting in lower pay. Sometimes the aim is not a compensation but an explicit incentive to women to ‘go back to their families’ to perform family chores and care activities, as was the Italian motivation for the incredibly favorable access conditions to retirement for women in the public sector (15 year of seniority, which created the so-called ‘baby retirees’!). As if ‘caring’ was natural for women and somewhat bizarre for men!

Then there are the differences in men/women psychological traits, such as the documented women's greater risk aversion and lower self-confidence, which typically lead women to more prudent portfolio choices and lower returns on their savings.

If this is the past and present of women's pensions, things are bound to change in the future, although perhaps at a lower than desirable pace. For better or for worse? The answer, as it is common in economics, is: ‘it depends’. It depends essentially on how younger women will perform, relatively to men, in the labor market with respect to the quantity, quality, and pay of their work. An increasing number of women have (more than) full qualifications to compete but true equality of opportunity is still a goal, not a realization (except perhaps in some Northern European Countries). Establishing parity of treatment in practice will reduce the gap if not cancel it entirely. Indeed, all over the world there have been pension reforms that have reduced their overall ‘generosity’ (and thus the cost to younger and future cohorts); reinforced the link, at the individual level, between contributions and benefits with the adoption of Defined Contribution (DC) kind of formulae and moved toward equality of retirement ages and actuarial pension formulae. Solidarity is taking the form of credited contributions to be paid out of general taxation for involuntary unemployment and/or absence from work for care activities, to be applied to both women and men; citizen's income is slowly taking the place of minimum pensions. Parental leaves are more common than just ‘maternal’ leaves and when the mother's presence is inevitable or recommended payroll taxes are (partially) covered through progressive taxation (not through contributions proportional to labor income).

Middle-aged women are in this transition while younger ones participate from the start to the new, more egalitarian, regime that is being built. They are better placed in terms of working opportunities, equal pay for equal work and equal chances of career progression; on the other side, they can count less on family ties and support, as marriage is less frequent and, on average, shorter due to divorce. They have to rely on themselves to achieve economic independence, for which an important role is participation in supplementary, fully funded pension plans, to make the most of portfolio diversification and take advantage of tax rates reduction. Although the general education female gap has disappeared in many countries (or, indeed, it has changed into a male gap), financial education, an essential factor for this positive evolution to occur, still exhibits a gap. Its advancement is also likely to help women to achieve more self-confidence and perhaps reduce the gender gap in risk aversion.

How rapidly will the pension gap be filled? First, women themselves have to assert their determination to play in a ‘level playing field’ and reject the practice of later compensations for earlier disadvantages (which does not mean that solidarity will not be needed, but that it should not be based on women being automatically considered the ‘weak’ gender). This is the path ahead, a still uneven path, full of cultural and psychological obstacles, which require appropriate policy measures, financial education programs and default designs to encourage the right choices. To make things even less certain are the consequences of the great COVID-19 crisis, which has had occupational, income and pension costs especially for women. It can be an opportunity to move ahead more swiftly or it can reduce the convergence to gender equality by creating room and scope for further paternalistic policies. It is up to women to demand the first as the only desirable and progressive path.