Research Of The Connection Between Growth And Debt Levels

There’s a lot of literature that suggests that there’s a negative relationship between growth and debt. The authors look at 17 OECD countries to see if this true. A key factor in this is that all of the countries have high levels of democracy. One problem with previous studies is that they don’t really factor for reverse causation, where low growth leads to high debt, and not vice versa. The authors combat this by using IV and 2sls, but introduce a new instrument, called Valuation Effect (VE). “Valuation effect brought about by the interaction between foreign currency debt and movements in the exchange rate” (Panizza and Presbitero, 2014). The main debt being studied is government. The authors also control for debt composition and the effective exchange rate.

Endogeneity Problem:

The authors have proposed a new instrument because they feel that there are some key problems with the instruments used in previous literature: “Instrumenting public debt with that of neighboring countries is problematic in the presence of global shocks and financial and real spillovers” and “The use of lagged variables is problematic because debt and growth tend to be persistent. ” These were the instruments that prior studies had used to show that a negative relationship between debt and GDP existed. VE is relevant except when its value is 0 (when a country has no foreign currency debt or a very stable exchange rate). France and Germany are prominent examples. However, VE may not be exogenous, as foreign currency debt can have a direct effect on GDP growth, and changes in the exchange rate can alter a country’s exports, also changing GDP growth. Therefore, these two factors have been controlled for.

Empirical Results:

The authors compare OLS regression with 2sls. Before controlling for endogeneity and using OLS, there’s a -1. 796% effect of debt on GDP. However, by using VE as in instrument for debt-GDP ratio, the negative effect disappears.

After controlling for currency composition and effective exchange rates (Panizza and Presbitero, 2014), the OLS regression indicates that there’s a -1. 563% effect. When we do 2sls with VE as an instrument, debt-GDP ratio has no effect and is statistically insignificant.

Interestingly, as stated above, all the countries are democracies, and none show any effect between debt and GDP. This finding is similar to what Kourtellos et al. 2013 found in their threshold study. However, we can’t say that the only reason that these countries don’t display a relationship between debt and GDP growth is because of their strong institutions; there could be many other factors (level of external debt, strong currencies, 1st world countries, etc. ) that could be the cause of this.

The authors use various checks for robustness. One of the more interesting checks was using placebo IV in their regression to see if the results change. They didn’t.

In conclusion, “We don’t find any evidence that, in the medium run, high public debt levels hurt future growth in advanced economies”