Budget Deficits
Party Political Premiums?

It is often argued that there is a direct relationship between government deficits and real interest rates. The conventional 'crowding-out' hypothesis suggests that this relationship arises because of competition for savings between the public and private sectors, which drives up real interest rates. This argument rests on an assumption that consumers do not take full account of the increased taxes they must pay in the future as a result of a higher budget deficit; otherwise private savings should rise to provide for these future tax liabilities. This would allow the increased government deficit to be financed at an unchanged rate of return to savings, so no crowding out will occur: reduced public savings are matched by increased private savings. Previous research on this link has found little, if any, evidence in its favour.

In Discussion Paper No. 79, Research Fellow Patrick Minford argues that government deficits do influence interest rates, but by creating risk premiums rather than through increased competition for savings. Political parties and their behaviour are an essential feature of the analysis, and Minford argues that the risk premium on real interest rates reflects the probability of 'inflationary default' which arises from 'party-political' differences. Higher levels of public debt, Minford argues, are associated with higher rates of inflation, a greater risk of inflationary default and thus higher interest rates.

Minford's analysis assumes that the economy is 'neo-Ricardian', in the sense that consumers fully discount future taxes. Consequently, when the government runs a deficit financed by issuing bonds to consumers, expected real rates of return on saving will not rise and crowding out (in the conventional sense) does not occur. Nevertheless, in Minford's model real rates of interest on nominally-denominated bonds do rise when the government deficit increases, because of the higher risk premium.

Minford's argument rests on differences between the policy behaviour of the two political parties. If elected, each party, 'left' or 'right' wing, will pursue policies designed to benefit its own supporters. Minford assumes that the left's policies will be more inflationary than the right's, because right supporters hold money-denominated financial assets, which can be expropriated by unanticipated inflation. This will benefit left supporters, who hold mainly human capital, Minford argues. This risk of post-electoral expropriation of nominal debt has to be compensated with a risk premium, which increases with the proportion of such debt in the private sector portfolio and hence the size of the bond-financed deficit.
Minford's analysis rests on the hypothesis that political parties differ in their policy behaviour; postwar evidence for the UK, West Germany and Sweden supports this assumption, he argues. The theory predicts that expected real interest rates on bonds are related to budget deficits and the level of public debt, and Minford finds that evidence for the US and UK from 1920 to 1982 supports this prediction. The theory also predicts an indirect relationship between the party in power and the level and variability of inflation. Minford finds some support for this relationship for the US and the UK, but the evidence is not strong. The theory also provides a rationale for the observed relationship between inflation and its variability.

Minford concludes that this analysis can help explain why government deficits raise real interest rates on bonds. The analysis also underlines the importance of political factors in macroeconomic analysis.


Interest Rates and Bond-Financed Deficits in a Ricardian Two-Party Democracy
P Minford

Discussion Paper No. 79, November 1985 (IM)