Have social security reforms shifted too much risk to individuals? The financial crisis suggests they might have

Monika Bütler

13 February 2009



The financial crisis comes right at a time when major reforms to social security systems around the world are evolving. These reforms were primarily initiated to tackle demographic transitions and resolve resulting fiscal imbalances. Measures taken or planned include strengthening employer-based, fully funded occupational pension schemes and full or partial privatisation of social security, often even including subsistence-level provisions. The reforms include more individual choice and greater individual risk.

In many countries, majorities of the populations are directly affected by the turbulence on the capital markets, even those who do not or did not privately own shares, at least in part due to a shift to more funded schemes. The world's pension funds have lost up to 30% or more of the value of their assets. Of the people in funded plans, the financial crisis hit those insured with individual schemes worst. While the older covered by collectively organised systems (such as those in the Netherlands and Switzerland) will feel little impact – perhaps even too little – many people with individual accounts have considerably less pension capital than they did the year before.

Yet pay-as-you-go old-age pensions are affected by the crisis as well – the recession-related decline in the contributions of workers lowers their revenue. Furthermore, these same systems will have to support and guarantee a subsistence-level income to all those who watched their private pension wealth dwindle in the financial crisis. In the end, governments might feel obligated to bail out under-funded plans to protect the retirees. This is a well-known, although somewhat forgotten, phenomenon. In many countries, the PAYG pension system is itself the child of a crisis. The US Social Security system was introduced in 1935 by Franklin D. Roosevelt as part of the New Deal.

The current crisis reminds us that the investment risk to retirement funds is significant. This is not in and of itself an argument against fully funded schemes. But perhaps it should make one wary of increasing individuals’ risk exposure. The crisis also reminds us that the government cannot remove itself from its involvement in social security as many had thought. Governments will have to reconsider their role in the provision of a basic income (protected against inflation and capital market fluctuations) and in setting up a judicious regulation of private and collectively organised pension schemes.

Old age security is insurance

Together with sickness and unemployment, age is one of life's major risks. A long life may result in an insufficient amount of wealth before death or a premature death due to a lack of security for survivors. In both cases, investment risks can be significant. As with unemployment or illness, the individual risks are substantially higher than the aggregate risk - a classical rationale for insurance.

Private insurance solutions can fail in hedging the financial risks of aging. Asymmetric information and different types of risks make insurance difficult, often just for those who need it most. Last but not least, national laws guaranteeing subsistence-level incomes (often anchored in Constitutions) represent an incentive to not insure privately against a long life and other contingencies associated with old age.

There exist two additional problematic issues for old age pensions. These are mainly due to the long span of time, several decades, within which an effective insurance must operate:

1) Not all risks are tradable

Macroeconomic risks, such as business cycles, inflation, and major crises, and other systemic risks, such as increasing life expectancy and standards of living, are not fully tradable risks. In the near absence of suitable instruments to hedge these contingencies, individuals may be exposed to more risk than desirable.

2) Individual choices are often imperfect

An extensive amount of literature reports behavioural anomalies with regard to provisions for old age. People don't voluntarily save enough for old age and their risks are insufficiently insured. People diversify incorrectly. They follow others or do what their employer recommends – even if this is often not optimal.

These factors, taken together, result in the state playing an important role in old age pensions. The impact of the financial crisis can help economists and policy makers assess the success of previous reforms and suggest routes to follow or to avoid.

Secure basic level of income in old age

In November 2008, Argentina nationalised private pension funds, which had lost almost half their value due to the financial crisis. Although the official justification "to protect the savings of pensioners and workers" may have been somewhat of an excuse, the political shift also shows how important it is to secure livelihood in old age – and how in a crisis it finds political support.

An efficient and parsimonious pay-as-you-go system is better suited than a funded pension system in securing pension wealth for two reasons. First, it insures all by redistributing in favour of the poor and families. Second, the pay-as-you-go system is in a better position to effectively insure individuals even in difficult times against the financial consequences of age without exposing the elderly to any major risks. This, however, is only the case if it only guarantees subsistence-level income and does not offer overly generous benefits.

It is interesting to note that even countries with no or very weak pay-as-you-go systems (such as Australia) ultimately rely on a substantial implicit pay-as-you-go component. National governments don't want to allow their elderly to starve, so states must fill gaps with means-tested and tax-funded benefits. The losses in wealth due to the international financial crises will encumber these countries with additional extraordinarily high expenses – these are trying times.

Better risk-sharing between generations

In the course of the ongoing reforms, many defined-benefit plans have been replaced by defined-contribution plans. But pure defined-contribution plans expose the insured in the pension funds to at least three types of changes in macroeconomic conditions. First, the value of the portfolio at the time of retirement. Second, through the annuitisation factor (e.g. indirectly through the yield curve) by which retirement balances are translated into life-long annuities. Third, inflation threatens to devalue the pension payments of the elderly. These risks have been greatly underestimated prior to the crisis. They can only be poorly hedged by individuals. Collective plans, despite having problems of their own, have proven more efficient in coping with these risks, as the examples of the Netherlands and Switzerland have shown.

The crisis suggests that a pure defined-contribution plan exposes individuals to too much risk – at a time when alternative income-generating avenues are also scarce. Nobody wants to return to the (rightfully discontinued) pure-defined benefit plans with their dependence on final pay in which richer individuals usually get a better deal. But there are alternative options available, such as defined-contribution plans with a benefit target. Here the pension is proportional to the pension contributions paid on average. With an appropriate regulation of the minimum accrual rate and the conversion rate, individuals remain more or less protected from macroeconomic risks.

Such a solution is tantamount to insurance between generations, which is easier to achieve in collectively organised systems. However, the crisis has also shown that the systemic risk of such intergenerational risk distribution has been underestimated, and the riskless benefits were too high in the past. Regulation must take this into account in future adjustments to the benefit components and reserves.

Choice with a safety net

Privatisation of old-age pension plans has brought individuals more choice. Not all solutions have proven to be wise. High returns on capital markets have lured individuals into not annuitising pension wealth or investing in risky products. Incorrect decisions not only threaten the individual but also have an effect on the state. By insuring a subsistence-level of income, the taxpayer bears the costs of sub-optimal decisions made by individuals with private pension plans.

The hope that, with appropriate education, people will make rational decisions with regard to investment or annuitisation of pension wealth has proven largely fictitious. Studies have shown, practically without exception, that information and transparency have little effect. What do have effects are appropriate default options and the simplicity of the decision.

Bad choices have little consequences in good times. But these are not good times. The crisis has taught us that pension reforms should also consider restrictions on choice when bad decisions threaten subsistence-level provisions in old age. Appropriate default options and requiring active decisions help individuals cope with bounded rationality. These reduce most insurees' exposure to risk and secure the budgeting process of the government as a reinsurer better without limiting the choices of well-informed and experienced individuals.

Global and local crises in recent decades have repeatedly illustrated the importance of wisely organising the pension system. As the crisis illustrates, reforms to the pension systems – under the pressure of demographic changes and in light of the rosy capital market outlook – have exposed individuals to too much risk. It would be good if retirement plans were once again primarily considered insurance rather pure investment decisions. Not least, this would put more emphasis on intergenerational risk sharing and a prudential regulation.

However, insurance between and within generations only works if the system is sustainably defined and financed. In this regard, the crisis has not changed anything.


Boeri, T, A.L. Bovenberg, B. Coeuré & A. Roberts, Dealing with the New Giants: Rethinking the Role of Pension Funds, International Center for Monetary and Banking Studies and CEPR (2006).
Bovenberg, A. Lans & Roel M.W.J. Beetsma, "Pensions, Intergenerational Risk Sharing and Inflation", Economica (forthcoming);
Bütler, Monika & Teppa, Federica: "The choice between an annuity and a lump sum: Results from Swiss pension funds", Journal of Public Economics (2007).
Carroll, Gabriel D., James Choi, David Laibson, Brigitte C. Madrian, & Andrew Metrick. "Optimal Defaults and Active Decisions." Quarterly Journal of Economics (forthcoming).



Topics:  Financial markets

Tags:  insurance, risk, safety net, social security reforms

Professor of Economics and Public Policy, St. Gallen University; Managing Director, Swiss Institute for Empirical Economic Research and CEPR Research Affiliate