The 90% public debt threshold: the rise and fall of a stylized fact

Abstract

We put the original Reinhart–Rogoff data-set, made public by Herndon et al. (2013), to a formal econometric test to identify public debt thresholds endogenously. We show that the nonlinear relation between debt and growth is not robust. Taken with a pinch of salt, our results suggest, however, that a negative association between central government debt and growth may set in at debt levels as low as 20% of GDP. Further (and greater) thresholds may exist, but their magnitude is uncertain. For general government debt, the threshold is considerably higher at about 50%. Country-specific estimates reveal a large amount of cross-country heterogeneity. For some countries including the United States, a nonlinear negative link can be detected at about 30% of GDP. For others, no nonlinearities can be established. Our results are a formal econometric confirmation that the 90% public debt threshold is not in the Reinhart–Rogoff data. But our results also seem to suggest that public debt be associated with poor economic performance at fairly moderate public debt levels. The absence of threshold effects or low estimated thresholds may not preclude the emergence of further threshold effects, especially as public debt levels are rising to unprecedentedly high levels.

Keywords:

• public debt
• economic growth
• nonlinearity
• threshold effects

JEL Classification:

• E60
• F30
• F40
• N40

Acknowledgements

The author would like to thank an anonymous referee and Bob Ford for useful comments and Mee-Lan Frank for excellent editorial support. This article is a revised version of the OECD Economics Department working paper No. 1055. The views expressed herein in no way represent those of the OECD.

Notes

¹ Égert (2012) matched data on central government debt obtained from the data appendix of Reinhart and Rogoff (2011) with real GDP growth rates available from the Barro–Ursúa macroeconomic data-set (Barro and Ursúa, 2012). The difference with the actual Reinhart and Rogoff data is that Égert’s (2012) data exclude Ireland and include Switzerland and that the data series used in Égert (2012) are longer.

² Available here: http://www.peri.umass.edu/236/hash/31e2ff374b6377b2ddec04deaa6388b1/publication/566/.

³ There are three differences in the data-set used in the article and the one used by Herndon et al. (2013). First, for France, the public debt series has a sudden break in 1978–1979 (8.8% in 1978 and 31.1% in 1979). We decided to use the series calculated from the separate nominal debt and nominal GDP series provided by Herndon et al. (2013). This leaves us with missing values from 1973 to 1977, but we avoid the abrupt break, probably due to a change in methodology/definition. Another difference is that Herndon et al. (2013) use the Greek public debt-to-GDP ratio starting in 1970, while the series can be computed from 1948 onwards. Finally, Herndon et al. (2013) exclude 1956 for the Netherlands, even though the observation is not missing from their background data file.

⁴ These results are not reported here but can be obtained upon request from the author.

⁵ Table A1 in the appendix provides more details on the time coverage.

⁶ Supplemental Data report results for the relationship between contemporaneous public debt and growth. The results do not differ too much from those obtained using lagged public debt.