Research Discussion Paper – RDP 8005 Social Security, Debt Finance, and Savings

Abstract

A recently revived topic of interest in the economics literature has been the effect of certain aspects of goverment behaviour on the economy. In particular, it has been argued that increases in social security pension payments and the use of debt rather than tax finance for government spending both tend to depress savings, and thereby capital formation. These assertions contradict a long-standing neutrality theorem that dates back to David Ricardo.

The essence of the early argument supporting neutrality was that individuals would only change their behaviour if the government action caused them to feel richer or poorer. Social security pensions involve a transfer from the young to the old. From the individual's point of view he receives when he is old approximately what he pays in when young. Similarly, debt involves a deferring of current taxation for future taxation. The interest on the debt, like social security, involves a transfer from one group to another and thereby no net leakage from the system. The neutrality argument rested on the assertion that a rational man would be aware of these transfer implications, realise that he and his descendants were no better or worse off, and therefore not alter his behaviour.

The modern explanation of neutrality, usually attributed to Robert Barro, is more complex. To begin with, it recognises that behaviour may alter even if life-time wealth is unaltered. The dominant impact of both social security and debt is an imposed rearranging of the individual's life-time consumption pattern. Neutrality will only occur if the individual can undo these intertemporal allocation effects through his own behaviour.

In this paper we show that if the economy behaves as a composite individual, then full neutrality is possible. Social security and debt both reduce consumption during youth and increase it during retirement; the rate of transfer being determined by the rate of population growth. However, if the economy has a system of operative intergenerational transfers (gifts or bequests) it is able to neutralise the reallocation effects by reversing the transfer, at a rate determined once again by the rate of population growth. This argument is only valid when we view the economy as a composite individual. The individuals themselves will not observe this perfect substitutability between social security, debt, and bequests due to corner solutions, illusions, and distribution effects. Whether or not the aggregate economy will display neutrality is an empirical question of the relative strengths of these forces.

The paper reports the results of estimating savings functions using annual and quarterly data for the 1960's and 1970's. On the basis of these equations we are unable to reject either of the neutrality hypotheses. The implications of these results are twofold: increases in the scale of our pension system should not be regarded as having a depressing effect on savings; and, the financing mix between taxes and debt is not important, in particular, there is no support for the argument that the public debt should be reduced by raising taxes.