Budget policy and the decline of national saving revisited

BIS Economic Papers No 33
March 1992

Introduction

The world saving rate declined significantly in the 1980s,' raising questions about the potential for a world saving "shortfall". The concern is that the level of saving and consequent rate of capital accumulation may not be adequate to support desired increases in standards of living over the medium and long-term time horizons. The Chairman of the Board of Governors of the Federal Reserve System in the United States recently stated, for example, that " ... there is no question that the decline in the US national saving rate has been costly, and that the recovery of that saving rate should be a national priority" (Greenspan (1991)). Similar views have been expressed by policy-makers and others in many countries and the global nature of the potential problem has also been emphasised (e.g. IMF (1991) and BIS (1991)). Moreover, the concern is heightened by the higher projected demands for capital in some countries -particularly the demand for new capital investment from eastern European countries, for rebuilding infrastructure damaged by the war in the Persian Gulf, to meet the needs of developing countries and to repair neglected infrastructure in several industrial countries. Although the present cyclical downturn may postpone the visible effects of a shortfall - a falling-off in investment and a rise in precautionary saving are typical at this stage of the business cycle - greater demands on worldwide saving may nonetheless arise over the medium term. One estimate suggests that the additional demand for saving might well exceed $100 billion in the years to come (Camdessus (1991)).

At the centre of the world saving shortfall discussion is the role played by government policy. In a purely accounting sense, the lion's share of the decline in national saving in many countries may be directly attributable to the fall in government saving in the 1980s, due both to the rise in budget deficits and a shift in government expenditure away from investment and towards consumption and transfers. However, budget actions also influence private saving through a variety of indirect channels, so that their net effect on national saving cannot be determined a priori. On one level, a direct linkage between private and public saving is often hypothesised. Proponents of the Ricardian equivalence hypothesis hold that a rise in the budget deficit due to a tax cut will be entirely offset by a rise in private saving - as households associate additional government debt issuance (due to the deficit) with higher future taxes, they increase saving to offset the anticipated liability. The conventional view, in contrast, holds that private saving will only partially respond to a rise in budget deficits - perhaps because households are not "forwardlooking" or altruistic, do not attempt to smooth consumption over time, or because of institutional factors, such as liquidity constraints -and the result is lower national saving.

In this paper we argue that the response of private saving to government budget policy is much more complicated than generally recognised, and that private saving cannot be expected to automatically offset the adverse effect on national saving arising from large budget deficits. Changes in the structure of government expenditure and taxes typically influence private decisions in a variety of ways, each of which may have an impact on private saving. The combination of these factors - both the direct demand on saving represented by government budget deficit finance and their indirect effect on private saving working through incentives created by changes in the level and composition of government tax and expenditure policy - may either reinforce or tend to offset each other in terms of the overall impact on national saving. Our analysis also questions whether there is a stable and reliable private saving response to government policy, either over time or across countries. This is partly due to the changing nature of government policy and institutional features of economies. But a more fundamental problem is our limited understanding of what determines private saving and the weak empirical support for even the most basic theories of saving behaviour. The analysis also considers several potential shortcomings in the measurement of deficits, and focuses on the growth of government contingent liabilities, implications for measures of the private saving offset to deficits and the potential intergenerational consequences.

In addition to surveying existing evidence, we present our own estimates of the response of private saving to government budget deficits for five major industrial economies, taking into account both the structure of the deficit and the composition of government expenditures. These estimates support the intuitive idea that the private sector response to changes in government budgets will depend crucially upon the underlying policy generating tile change and that, in general, a strong private saving offset to a decline in government finances cannot be relied upon to support national saving. In particular, we find that when the deterioration in government finances is attributable to expenditure increases, rather than tax reductions, national saving is reduced. Since most of the fall in government saving during the past two decades is associated with the growth of government expenditures, it follows that budgetary policies have been a primary cause of the fall in national saving. This argument is strengthened by the analysis of government contingent liabilities. We find that conventional empirical estimates of the private saving response to budget deficits based on national income statistics greatly exaggerate the extent of private saving offset. Most importantly, the rise in massive government contingent obligations, typically off-budget and unfunded, has led to a significant underestimation of the growth of total government liabilities, implying a significant "intergenerational burden" on future generations. This is additional evidence supporting the view that the deterioration of public finances represents a significant drain on national saving and imposes real economic costs on future generations.

In the first section we review some of the broad trends in private and public saving over the last three decades. In section II we discuss the basic theoretical issues, review the empirical evidence and present estimates of the private saving offset to government budget deficits for five major industrial economies. In section III we present estimates of an expanded model taking into account both the structure of the deficit and the composition of government expenditure. We also discuss the plausibility and stability of the underlying model of saving behaviour and compare our results with other empirical studies. In section IV we present evidence on the growth of government contingent liabilities and discuss its implications. A concluding section draws some policy implications from the analysis. Two technical appendices present some of the theoretical arguments more rigorously, describe the empirical methodology and present some of the preliminary statistical results.