Revisiting the bi-directional causality between debt and growth: Evidence from linear and nonlinear tests
Revisiting the bi-directional causality between debt and growth: Evidence from linear and nonlinear tests
We revisit the bi-directional causality between public debt and the rate of GDP growth for 10 EMU countries alongside the US, UK and Japan, over sample periods spanning from 1970 up to 2014 whilst accounting for the nonlinear properties of both the individual time series, and their relation in both directions. Our results indicate that the causal relation between debt and growth, in either direction, is weak at best. For most of the countries in our sample, we find no robust evidence of a long-run causal effect using bi-variate Granger causality tests. Bi-directional causality is detected only for Austria, while for France, Luxembourg and Portugal, causality runs solely from debt to growth, but the estimated effects are very small. In Finland, Spain and Italy, Granger causality (from growth to debt in the former two and debt to growth in Italy) appears to be present in the short-run. Our findings cannot be taken as evidence that a high level of public indebtedness is not risky for the economy or as invalidating hypotheses postulating effects in either direction in the relation between debt and growth.
Public debt, Economic growth, Granger causality, Nonlinear causality, SYS-GMM
55-74
De Vita, Glauco
002fc6bf-e5ed-4a13-8993-0ce5e1fc2005
Trachanas, Emmanouil
4d570d27-0a47-46e6-a333-0e66f4700b05
Luo, Yun
2ac0f228-573d-43e7-b309-1529b6f3d174
May 2018
De Vita, Glauco
002fc6bf-e5ed-4a13-8993-0ce5e1fc2005
Trachanas, Emmanouil
4d570d27-0a47-46e6-a333-0e66f4700b05
Luo, Yun
2ac0f228-573d-43e7-b309-1529b6f3d174
De Vita, Glauco, Trachanas, Emmanouil and Luo, Yun
(2018)
Revisiting the bi-directional causality between debt and growth: Evidence from linear and nonlinear tests.
Journal of International Money and Finance, 83, .
(doi:10.1016/j.jimonfin.2018.02.004).
Abstract
We revisit the bi-directional causality between public debt and the rate of GDP growth for 10 EMU countries alongside the US, UK and Japan, over sample periods spanning from 1970 up to 2014 whilst accounting for the nonlinear properties of both the individual time series, and their relation in both directions. Our results indicate that the causal relation between debt and growth, in either direction, is weak at best. For most of the countries in our sample, we find no robust evidence of a long-run causal effect using bi-variate Granger causality tests. Bi-directional causality is detected only for Austria, while for France, Luxembourg and Portugal, causality runs solely from debt to growth, but the estimated effects are very small. In Finland, Spain and Italy, Granger causality (from growth to debt in the former two and debt to growth in Italy) appears to be present in the short-run. Our findings cannot be taken as evidence that a high level of public indebtedness is not risky for the economy or as invalidating hypotheses postulating effects in either direction in the relation between debt and growth.
Text
JES accepted paper on 9 Feb 2018 FDI Export Nigeria paper
- Accepted Manuscript
Text
IJFE
- Accepted Manuscript
Restricted to Repository staff only
Request a copy
Text
Revisiting the bi-directional causality between debt and growth: Evidence from linear and nonlinear tests
- Version of Record
Restricted to Repository staff only
Request a copy
More information
Accepted/In Press date: 1 April 2016
e-pub ahead of print date: 11 February 2018
Published date: May 2018
Keywords:
Public debt, Economic growth, Granger causality, Nonlinear causality, SYS-GMM
Identifiers
Local EPrints ID: 437970
URI: http://eprints.soton.ac.uk/id/eprint/437970
ISSN: 0261-5606
PURE UUID: 900a96a3-8535-4089-bd1f-59f2fb66e5b2
Catalogue record
Date deposited: 25 Feb 2020 17:30
Last modified: 17 Mar 2024 05:20
Export record
Altmetrics
Contributors
Author:
Glauco De Vita
Author:
Emmanouil Trachanas
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
