This book provides a wealth of information and analysis relevant to social security reform. The authors survey the demographic and fiscal background and their implications for retirement income distribution and employment, in the US and other industrialized countries. Drawing heavily on OECD data and research, they cover topics such as government pension liabilities, household pension wealth, and the impact of fertility rates and immigration. A discussion of retirement age – its causes and effects – is a focal point of the book. Other topics covered include poverty alleviation, inter-generational transfers, and public sector employee pensions. Extensive tables, figures, references, and illustrative cases from various European countries are especially useful.
The authors' analysis leads to a set of reform proposals for the US: raise future retirement ages by automatically linking them to longevity increases; cut benefits of high earners by offsetting part of the benefits they receive from tax-advantaged retirement saving accounts; and increase revenues for social security to fill in the remaining gap (they don't specify the revenue source). They also recommend supplementing social security with add-on quasi-mandatory retirement saving accounts with automatic enrollment and opt-out provisions. These accounts would be low-cost, collected through the government, and would have limited and privately-managed investment options with age-adjusted portfolios as the default.
Overall I generally agree with these proposals. The emphasis on automatic adjustments is welcome, especially as the response to persistent situations including on-going longevity growth. In my view, this is the best way to take difficult decisions out of the political arena and get fast – and therefore more timely and less costly – actions. I am not sure that cutting social security benefits proportional to tax advantages received from retirement saving accounts is the best way to go, as this might deter such saving. Further, high earners who lack such accounts may find other (more effective) ways to exploit tax advantages. The authors do not propose cutting spousal benefits to middle and upper-class wives, although the current US situation is difficult to justify on equity or efficiency grounds. And I would anticipate that the add-on accounts will eventually prevail. One problem with the automatic enrollment plus opt-out approach (which has become popular as a way to get the benefits of taxation without the bad political connotations) is that some workers may not understand what is happening and may be dismayed later on when they find they cannot take their money out to cover periods of unemployment or medical emergencies. Also, in view of stagnating wages at the lower end of the income distribution, quasi-mandatory contributions together with the increasing cost of health insurance may lead to cuts in standard of living unless these are partly financed by matching grants from government revenues (which would pose other problems).
A strength and weakness of the book is its broad scope. The fact that it covers many topics and sub-topics in detail is useful, but at times one loses sight of over-arching themes, missing the forest for the trees. Some passages are convoluted and those not already familiar with the literature may have a hard time understanding what they are saying. For example, in their discussion of progressivity, it is not always clear whether the authors are discussing annual or lifetime benefits. Since one's expected lifetime will be highly correlated with one's socio-economic status, many systems that appear progressive based on annual benefits turn out not to be progressive when measured by lifetime benefits.
The book also frames its discussion of benefits to public sector employees as a distributional issue, in that public employees generally receive higher old-age benefits than do private sector employees. But if one thinks these benefits are ultimately paid for by the worker in the form of lower wages, this is not a distributional issue but rather an efficient compensation issue. The right question is which form of compensation leads to the best public staff and the most sustainable level of state and local services? Perhaps the public sector has paid in the form of higher benefits and lower wages because benefits are less visible, therefore less likely to be attacked politically. But this introduces several inefficiencies: it impedes inter-sector mobility for workers already hired, and some potential future high-quality public sector employees might choose to work elsewhere, preferring to receive higher wages. Also it impedes good decision-making, as it initially allows states and localities to underfund and therefore hide the true cost of their compensation programs. As this cost illusion is revealed, it may eventually lead to unmet expectations – of benefits for retirees, public services, and tax rates for state residents.
The authors' discussion of occupational plans is useful, making the important point that historically these have mainly benefited high earners. Further, currently we see a shift from defined benefit to defined contribution plans, not an overall reduction in coverage. The authors attribute this shift to an increase in longevity and volatility of market returns. By contrast, I would argue it is not simply an increase in these costs but due to more basic forces such as plan maturation and regulation, which force employers to recognize existing costs. Initially, the plans were not mature and had few retirees; this kept cash flow costs low, especially since full funding of future benefits was not required. The plans also had high vesting requirements, which meant they would never have many retirees. Costs therefore appeared low because of an illusion on the part of the employers that they would not have to pay much, and on the part of workers that they might actually get these benefits despite the lack of vesting and funding. As plans matured, hence had more retirees, and as regulations increased, forcing employers to acknowledge their pension promises and the high cost of covering longevity and investment risk (more recently, even putting the pension liabilities on their books), employers realized that simply paying wages as well as defined contribution pension contributions would cost less but allow them to attract an equivalent group of workers (given worker myopia and turnover).
I found the chapter on individual account systems in Latin America disappointing and sometimes misleading. The authors point to the “dire” problem of old age poverty in Chile, but the data they present show that poverty in households with elderly members is substantially lower than poverty in other households. They also claim that the new (2008) public supplementary benefit that goes to retirees with small private accounts subsidizes and incentivizes their contributions, but in fact it is an imputed 30% tax on private contributions since it represents a phase-out of the basic benefit that goes to retirees with no private accounts. Their discussion of Mexico also misses the basic point that current retirees, as well as all retirees in the foreseeable future, will be in the old system because of grandfathering provisions. Hence any empirical discussion of current benefit size and distribution is really about the old system, not the new one. Coverage and contribution density issues are indeed important in Latin America, but this is not specific to DC systems. Instead, this problem is found in any system in which benefits are linked to contributions. And in countries where many workers are self-employed or in the informal sector, governments often cannot effectively collect payroll taxes. If such countries want to provide benefits to everyone, they must finance these benefits by some general revenue source – but that is difficult because general revenues are also difficult to collect and face many competing demands. So it boils down to a broad issue of fiscal capacity and priorities, rather than a narrow pension issue.
The discussion of the notional defined contribution (NDC) system in Sweden is useful. Yet here too it would have helped to have a succinct statement that the NDC (which is a pay-as-you-go scheme) is intended to achieve the labor market incentives of a close linkage between benefits and contributions without incurring the transition costs of a shift to a funded system. The NDC plan can also reduce future benefits implicitly, without making an explicit statement along these lines. However, the fact that, over the long term, the imputed rate of return in an unfunded NDC system is lower than the market rate in a funded system means that it remains unfair actuarially, so evasion may be a problem in countries without effective tax collection systems. Overall, this is an extremely useful reference book on the myriad of issues around pension reform and on how these issues have been and might be handled in European countries as well as the US.
