Sovereign indebtedness and financial and fiscal conditions

ABSTRACT

We empirically assess the magnitudes of sovereign indebtedness responses for a sample of 123 Advanced and Emerging Market Economies, between 1980 and 2018, taking into account the changing characteristics of financial markets, notably the Global and Financial Crisis. Our results show that when the financial conditions are more stressful, for instance, higher yield spreads or a heightened degree of financial stress, fiscal authorities use more actively their primary balance to reduce sovereign indebtedness, which is not the case when financial market conditions are more benign. This is notably true for the case of Emerging Market Economies sovereigns, who most likely then struggle more to fund themselves.

KEYWORDS:

• Sovereign indebtedness
• panel data
• financial stress
• global financial crisis
• emerging markets

JEL CLASSIFICATION:

• C23
• G15
• H63

Acknowledgments

The opinions expressed herein are those of the authors and do not necessarily those of their employers. Thanks go to the editor and two anonymous referees for useful comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Some researchers have examined financial sector and crisis-related determinants of sovereign bond spreads (see, e.g. Ebner (2009) and Dailami, Masson, and Padou (2008)).

² See Leeper (1991), Sims (1994) and Woodford (1995).

³ Alternatively, we also employed the IMF’s WEO output gap which yielded qualitatively similar results, even at the cost of a smaller number of observations. In fact, the bivariate correlation between the common set of WEO and HP-generated output gaps was 85% in our sample. The reason why the WEO output gap was not used as the baseline relates to the fact that these estimates are problematic in a cross-country analysis as they are based on methodologies that vary by country desk. As such, differences in gap estimates across countries are also due to the different methodologies used, rather than differences in economic fundamentals. In addition, by applying a filter we maximize the total number of observations available in a consistent and uniform manner.

⁴ The FSI data are only available until 2014. We use the spreads data ending also in 2014 for comparability purposes.

⁵ The FSI variable was retrieved from Balakrishnan et al. (2009) extended to 2014. This index represents an important improvement vis-à-vis zero-one binary variables, as it measures the intensity of stress and it is not ambiguous regarding ‘near-miss’ events. Moreover, it covers both abrupt rises in risk and uncertainty, concerns about the health of the financial sector, large swings in asset prices and sudden drops in liquidity. Bond yields come from Datastream and Bloomberg and the spread was computed vis-à-vis the US benchmark.

⁶ Bellas, Papaionannou, and Petrova (2010) findings indicate that financial sector vulnerabilities appear to be a crucial factor in explaining movements in spreads in EMEs.

Additional information

Funding

UECE is supported by FCT (Fundação para a Ciência e a Tecnologia, Portugal); National Science Foundation Portugal.