Economists have long argued over whether inflation is generally conducive or detrimental to growth, not least in debates on long-term development in Latin America. While monetarists generally consider price stability a prerequisite for growth, structuralists contend that monetary or fiscal restraint will result in unemployment and slow growth. So far, no clear conclusions have emerged from empirical studies of this issue, partly because of the difficulty of subjecting these complex relationships to conventional econometric analysis.
In Discussion Paper No. 375, Research Fellow Thorvaldur Gylfason reconsiders the relationship between inflation, economic growth and external debt. Due to the complexity of this relationship and the lack of adequate data on many countries, Gylfason does not attempt to estimate a conventional parameterized model. Instead, he uses a simple framework of the simultaneous determination of inflation and growth.
The empirical analysis focuses on cross-section data on inflation, debt and growth in 24 economies (in Latin America, Africa and Europe) with an average inflation rate of 86% during 1980-5, and 13 economies (predominantly Asian or oil-producing but also including Japan and Germany) with an average inflation rate of 3%. Having identified patterns in the data, he investigates their statistical significance using simple non-parametric techniques that do not require the samples under study to be normally distributed or to have equal variances.
One of the hypotheses tested is that before the supply shocks of the 1970s, when demand variations were pronounced, inflation and growth would be positively correlated. The results do not generally support a causal connection between high inflation and high growth during 1965-73. Though the association of high inflation with high debt as a proportion of GDP did pass less stringent tests of significance for some sub-periods, Gylfason generally fails to find a significant relationship between these variables over the whole period 1965-85.
The tests do, however, reveal significantly slower growth and higher debt in the high-inflation economies in various sub-periods since 1973. This is consistent with the view that the supply shocks of the 1970s were important in the simultaneous determination of inflation and growth, or that rapid inflation slowed growth by discouraging saving, or that slow growth led to increased inflation via tax erosion.

The results do not exclude other possibilities: that statistically significant differences among average growth rates in the two groups of countries are due to political stability or rainfall, in LDCs, or that growth and inflation are jointly influenced by institutional and structural phenomena. Nevertheless, if inflation is partly to blame for the average medium-term growth differential reported in the paper, Gylfason concludes, then the real costs of inflation over time may be quite high for both individual countries and the world economy as a whole.

Inflation, Growth, and External Debt: A View of the Landscape
Thorvaldur Gylfason

Discussion Paper No. 375, February 1990 (IM)