Research: The need for retirement bonds
Government-issued retirement bonds with a GDP-related coupon would help redress the savings imbalance by increasing government investment and mitigating the effects of quantative easing on pension funds, Jean Frijns, Theo van de Klundert and Anton van Nunen write
The European economy is stagnating. One of the causes is ageing, and that cause also indicates that this stagnation could be long-term. In the economic literature this tendency is known as secular stagnation. Ageing leads to additional saving, which in an ideal world would be absorbed by the private sector. Labour also becomes relatively scarce and capital superfluous, which will lead to a higher capital-labour ratio (capital deepening) through a change in factor price ratios. However, in the real world this tendency is not strong enough, partly because the need for physical capital decreases (less demand for new housing, changes in sector structure to more services) and partly because capital has taken on a more public character (fundamental research, IT infrastructure, etc). On top of that, governments slow down their investments, forced by budgetary rules. A third factor is that investments in small and medium-sized companies lag as the financing banks shorten their balance sheets.
The consequence of this stagnation is low interest rates. The policy reaction in the form of quantitative easing lowers interest rates further. Funded pension systems are hurt disproportionally badly and it is regrettable that precisely these pension schemes suffer from the adverse effects of quantative easing (QE), so collective savings do not decline, they increase.
A possible solution is the transformation of part of the funded system into a pay-as-you-go system. For the time being, this is only a strict theoretical solution: a large unfunded system at national level does not sit well with existing institutions. We think introduction of that change is unlikely.
We therefore propose an alternative, which has some of the characteristics of an unfunded system as it offers a stable albeit low return but is also directed towards increasing government investment. We propose that governments issue retirement bonds (RBs) in which part of the pension savings could be invested. The issuing governments would guarantee coupon payments and redemption and the yield would equal nominal GNP growth. Revenues would be allocated to government investments to realise the targeted macroeconomic capital ratio as an answer to ageing.
It is important that financing of RBs should not form part of the existing rules for government deficits. In a sense, these RBs are an alternative to the Juncker plan. Hence our advice to keep RBs separate from the Maastricht requirement to restrict government deficits and public debt levels.
Before we dive deeper into the macroeconomic consequences, it is conducive to paint the microeconomic perspective. We start with the deviation that RBs represent from capital market pricing and the consequences for the pension funds. The peculiar character of RBs is that the return is related to GDP growth. They represent a deliberate deviation from the dicates of capital markets and would be a vote for financial subsidiarity. The motivation for this is found in the institutional area. Countries do have the liberty to model part of their pension system on a pay-as-you-go basis whereby the government takes over part of the capital markets’ savings function. Furthermore, future liabilities on government finances are not part of EU budget rulings. The proposed RB system combines elements of pay-as-you-go (the government as an alternative for savings through the capital markets) with those of long-term budget sustainability by using the proceeds for raising the macro capital ratio.
Nevertheless, it can be asked whether it is wise to guarantee a return that clearly deviates from market practice and, more specifically, whether it is not counterproductive to have a positive return in times of a savings glut and negative interest rates.
In general, the answer to the latest question is affirmative. However, within the financial framework of pension funds, a lower capital market interest rate leads to a lower compound rate and thus to higher collective savings. RBs mitigate this pro-cyclical reaction. However, this does require that the use of RBs would be strictly limited to pension savings. We also imagine that rationing would be necessary.
Many funded pension systems are locked into a tight regulatory framework in which the general characteristic is extreme susceptibility to the nominal rate of interest. This is a weakness because pension liabilities or ambitions are not nominal but real in nature, and because the long-term instability of long-term interest rates renders the overall pension system unstable. When issuing RBs on a substantial scale it would be wise to base the compound rate on the guaranteed RB return. That will create more stability and will probably be favourable in the short term for pension funds’ solvency within the existing nominal framework.
An assessment from a macroeconomic view will have a short and a long-term perspective. The short-term consequences for nominal GBP depend on the effects on savings and (government) investment. We have already indicated that under current regulation pension funds tend to react perversely to a decrease in the interest rate: they will save more instead of less. Assuming that the introduction of RBs is seen as an opportunity to base the compound rate for pension liabilities on the interest rate on RBs, their introduction could lead to a less pro-cyclical development of collective savings. In concrete terms, in the present phase, introducing a guaranteed return as the compound rate will lead to lower or at least constant collective savings. As for the size of investments, this depends on alternatives for pension savings. Within the framework of pension funds, RBs would be treated as risk-free investments and, therefore, a substitute for investments in government bonds. The difference to government bonds is that RBs will directly be followed by additional government investments . Incidentally, a similar effect could be obtained by the Juncker plan, provided it is successful.
The long-term consequences for the sustainability of government finances are important. For reasons of simplicity, we depart from the idea that the economy, after a transition phase of ageing and capital deepening, will show a tendency toward a stable equilibrium growth rate which is low (in the range of 1% to 1.5% per annum). Government investment, as far as it is financed by RBs, rises annually with economic growth, and the same goes for the overall investments by pension funds in RBs. Interest payments on RBs are to be paid from (higher) tax revenues. We presume that the effect of this higher macroeconomic capital ratio will be more than sufficient to generate these extra tax revenues. The macro tax burden, however, will rise slightly, a reflection of the enlarged role the government will play in economic capital formation, which is a necessity caused by ageing.
Jean Frijns is a former CIO of ABP. Theo van der Klundert an academic at Tilburg University and Anton van Nunen a former director of strategic pension management at Syntrus Achmea