The industrialized world must face the consequence of a rapidly ageing population. The demographic dynamics poses a challenge to two key institutions of the modern welfare state: public health and pensions. The demand for high-quality services and generous pension earnings contrasts with the reality of a growing share of the population that retires becoming dependent on a declining share of workers. According to the population forecasts of the United Nations, the ratio between the population over 65 and the population in age 20-64 in Switzerland will grow from today’s 29% to 55% by 2050. The situation of Switzerland is problematic even in comparison with other industrialized countries: for instance, the old-age dependency ratio of the United States is estimated to be «only» 41% in 2050.
In principle, there could be three ways out of the demographic crisis that do not entail either raising taxes or cutting pension benefits. The first option would be to allow more immigrants. Immigrants typically arrive at an early stage of their working age, and have a higher fertility. A number of studies show immigration to have a large positive quantitative effective on government budgets. However, this option raises well-known resistance in the Swiss society. The second option would be to increase the retirement age. This may help, but one must respect the legitimate demand for flexibility and free choice in the rules of exit of the labor force. Finally, one might hope in faster productivity growth. However, in the current economic conditions no productivity boom is in sight.
«Pensions Reform 2020» is insufficient
Given these limitations, one must resort to changing the rules of the pension system. Switzerland is in better shape than are most Western countries as far as the implicit debt hanging on the future generations is concerned. Yet, its current pension system is not financially sustainable: by the current rules, the present value of future benefits exceeds the present value of the future revenues. It is therefore necessary to find an affordable and fair way to split the burden of the adjustment across different generations. The proposal «Pensions Reform 2020» limits the generosity of the second pillar (Berufliche Vorsorge) by reducing the conversion rates from 6.8% to 6%. This will force employers and employees to increase contributions in order to maintain the benefits at the current level. It also proposes a modest increase in the retirement age of women, while introducing more flexibility in the retirement rules. Finally, it grants an increase (moderate, but not inexpensive) in the AHV benefits, to be financed by an increase in the VAT. Employers and union dislike the reform for opposite reasons.
The reform proposal may well represent a reasonable political compromise. However, it fails to address a fundamental conceptual ambiguity. A pension system can be based on two principles (or a combination of them). The first is that of a transfer-based pay-as-you-go (PAYGO) system whereby workers (and/or their employers) are taxed and the proceeds are transferred to the retirees. The virtue of this system is that it yields intergenerational insurance – unlucky generations hit by negative shocks can be rescued by the transfers provided by future generations. Rules are often designed so as to provide redistribution also within generations, for instance, by guaranteeing a minimum level of pension benefits. This principle inspires the so-called first pillar of the Swiss system.
The implicit rate of return of a PAYGO system is the sum of the working population and productivity growth rate. In periods when these were higher, the PAYGO system could afford to pay high benefits.
Unfortunately, in the current times of low growth and fast-ageing population, generous benefits could only be sustained by increasing progressively the tax burden. Doing this would in turn distort the economy, and likely reduce further productivity growth. How generous can the system afford to be, then? Demographic dynamics (except for immigration rules that are subject to the voters’ whims) are easy to predict. On the contrary, productivity growth is uncertain. The current approach appears to be that of revising periodically the rules. This is far from ideal, as it gives the politicians in office an incentive to procrastinate any radical intervention, leaving open uncertainty about what will happen in the medium and long run.
In favor of self-equilibrating rules
A different approach would be that of designing self-equilibrating rules. This is the approach followed by Sweden in its 1994 reform. Rather than committing to a particular defined level of future benefits, the Swedish state pension agency makes forecasts based on different productivity growth scenarios, and consistent with an intertemporal balancing of the budget. Workers receive annually an updated version of their prospective pension under the alternative scenarios and can adjust their saving decisions accordingly.
The alternative (or complementary) principle is that of a fully-funded pension system. In such a system, workers’ contributions are not transferred to retirees, but are held as individual accounts invested in financial or real estate markets. In other words, pensions are a form of forced savings. The reason why the state might desire to force people to save through a funded pension system (as opposed to let them decide privately how much and in which form to save) is debatable. The typical arguments involve forms of paternalism: people may be unable to plan, or be financially irresponsibly, fall into poverty and then resort to other forms of public transfers. Leaving aside its deep rationale, this system has the advantage that it is automatically balanced. When the rate of return in capital markets falls, future benefits are reduced accordingly – as it happens with standard financial products such as investment funds.
This principle inspires the second pillar of the Swiss pension system, although its application is highly imperfect. The Swiss legislator appears to seek to «have the cake and eat it». Minimum guarantees on the rate of return of capital and the so-called conversion rates, often set with disregard of the evolution of market rates, are inconsistent with the principles of a funded system. Funded systems impose no burden on future generations, but are intrinsically riskier for the beneficiaries. This risk cannot be avoided while keeping the funded system in place. The legislator should scrap all political rules, and let payment be determined by market rates.
Swedish reform could serve as a model
One might worry about the incentives and the efficiency of the various Pensionkassen if these are not required to deliver a minimum return. Again, the Swedish reform of 1994 could serve as a model. In Sweden, the contributors can choose their preferred pension plan among different products with different risk and return characteristics. Competition (together with monitoring) is expected to provide the necessary discipline against the risk of mismanagement and abuses. The state provide savers with a relatively safe default option, for workers who wish to minimize risk.
In summary, the current Swiss system suffers from the adoption of hybrid principles within each of the two pillars. These creates in the public opinion the false impression that a certain pension level can be guaranteed irrespective of external circumstances. This expectation must then be fulfilled through periodical bail-outs. However, adjustments will become harder and harder as the old-age dependency ratio increases. It would be much better to take the bull by the horns. The return to the first pillar should adjust automatically in response to changes in productivity growth and demographics (especially, immigration rules). The second pillar should offer unadjusted market rates and allow people more flexibility in managing their investment strategy. The only important remaining political decision would then be the relative size of the two pillars. This choice is political, and economists can at best provide a reasonable measure of the trade-off between efficiency and redistribution.