This in a stimulating monograph which explores the relationship between the evolution of public pension (social security) programs and fertility choices made by families. Drawing on the classic Samuelson analogy of pay-as-you-go (PAYG) pensions as “social contracts” between generations, Cigno and Werding argue that families represent the most basic form of intergenerational contract. The intrafamily decision as to how many children to have, and as to how resources are distributed between generations, parallels decisions made in PAYG pensions. But the introduction of a pension program directly affects the incentives to “invest” in children, so that the observed combination of low fertility and highly developed public pension programs in OECD countries is intrinsically related.
Of course, most countries that provide pensions also provide a degree of child support; however, the authors' attempts to quantify these various financial flows suggest that the impact on life-cycle income of pension transfers far outweigh in value those associated with children. Since investment in children has a greater return to society than large transfers of welfare payments to the elderly, the authors have been in the forefront of those who argue that tax systems should to some extent remedy the balance whereby fertility incentives are diminished at the expense of, say, inducements to retire early.
The novelty of the present book in the context of this general argument is to examine the design of public pension systems. At the heart of its argument is a model that attempts to differentiate the more redistributive public pension program of the “Beveridge” form from the “actuarially-balance” program of the “Bismarck” form. The authors seek to argue that the former distorts output decisions and reduces saving rates, whereas the latter only causes distortions insofar as there is an “implicit tax” arising from PAYG from full funding. Moreover, they argue, the outcome (Section 5.2) is that a “Beveridge” program reduces saving and encourages people to have fewer children, relative to a Bismarck program – an idea also formalized recently by one of the authors (Cigno, Reference Cigno2008).
I have no problems with the idea that alternative pension designs have differing impacts on output and saving incentives, although my own evidence (Disney, Reference Disney2004, Reference Disney2006) is certainly more consistent with the “traditional” view that “Beveridge-type” pensions are more likely to induce output (labor supply) distortions whereas “Bismarck-style” programs are more likely to crowd out household saving in the form of private pensions (as illustrated by the dearth of private pensions in countries with Bismarck-style regimes). It is certainly true that redistribution to the lifecycle poor implicit in a Beveridge-style regime reduces the incentive to have children as a future “alternative” to state benefits for those groups, but it is marginal tax rates and average replacement rates that matter in all this. Thus it is not clear to me a priori why the distortions in terms of family size are greater in the Beveridge-style setting, notwithstanding the authors' model.
All this cries out for some empirical analysis, in addition to the stylized simulations for individual countries, primarily Germany, reported in this book. Although the Japanese pension program is arguably somewhat “Beveridgean” in design; if I were to investigate some stylized associations across OECD countries, I would tend to the view that fertility declines have in practice been steepest in countries that have adopted more “Bismarck-style” pension regimes, rather than the reverse. Of course, fertility declines are associated with a variety of factors (culture, women's labor market participation, and so on), while the relative development of public pensions also has much to do with other economic factors, such as the development of capital markets and the extent of household liquidity constraints. The whole pattern across time and countries in fertility trends and in the evolution of pension programs is complex and multifaceted and one model cannot capture it in its entirety.
This leads to a last thought. Since, much of the value of social transfers to children is regulated through, for example, the provision of public services such as schooling and pediatrics, different “models” of intergenerational family transfers across societies are surely summed up by differences in discretionary behavior by families. And the most obvious indication of such behavior is bequests (inheritances). That neither word appears in the index, in a book on children and pensions, is somewhat surprising. Although data on bequests is notoriously hard to find on a consistent cross-country basis, it would seem that such data would offer a rich source of testable hypotheses as to the relationship between institutional features such as pension design and tax structure on the one hand, and attitudes to children and grandchildren on the other.
Overall, however, the authors have presented a stimulating agenda for future research, and it is hoped others will rise to the challenge.
