Growth beyond the debt thresholds

This empirical study has two objectives both of which are directly related to nonlinear effects of debt on growth: The main objective is to assess the levels of public debt-to-GDP and total external debt-to-GDP ratios above which economic growth is impaired. The second objective is to examine the debt overhang effects on the sources of growth. A large panel data set of developed and developing countries over the period of 1970-2009 is employed to estimate the threshold levels for public and total external debt. The dynamic panel threshold method developed by Kremer et. al (2010) is applied to debt-growth nexus for the first time to estimate the debt thresholds. The method estimates threshold values from the data rather than imposing arbitrary values to study the debt threshold effects as had been done in the literature until now. The level of public debt above which the growth starts being a burden is 69% for high income OECD countries, 47% for middle income countries and 30% for low income. The model is modified to distinguish between the marginal and average debt overhang effects on growth. For high income OECD countries, once public debt-to-GDP exceeds the threshold, each additional unit of public debt lowers growth rate significanlty by about 0.1 percentage point. In middle income countries, however, the public debt overhang effect is observed through a significant slow down in average growth rate rather than a marginal impact on growth. For external debt, the threshold is around 80% for high income OECD countries, 50% for middle income countries, and 70% for low income group. Once these thresholds are exceeded, all countries face with a considerable decrease in their growth rates but the impact of each additional unit of external debt above the threshold is not significant. Therefore, countries' trend-growth is not effected as they accumulate more external debt beyond the threshold. For both public and external debt, the negative realtionship between debt and growth is mostly due to the decrease in capital stock accumulation for high debt levels rather than TFP growth.