Journal Article
No. 2007-14 | December 13, 2007
Testing for Breaks in Cointegrated Panels - with an Application to the Feldstein-Horioka Puzzle


Stability tests for cointegrating coefficients are known to have very low power with small to medium sample sizes. In this paper we propose to solve this problem by extending the tests to dependent cointegrated panels through the stationary bootstrap. Simulation evidence shows that the proposed panel tests improve considerably on asymptotic tests applied to individual series. As an empirical illustration we examined investment and saving for a panel of European countries over the 1960-2002 period. While the individual stability tests, contrary to expectations and graphical evidence, in almost all cases do not reject the null of stability, the bootstrap panel tests lead to the more plausible conclusion that the long-run relationship between these two variables is likely to have undergone a break.

JEL Classification:

C15, C23



Cite As

Francesca Di Iorio and Stefano Fachin (2008). Testing for Breaks in Cointegrated Panels - with an Application to the Feldstein-Horioka Puzzle. Economics: The Open-Access, Open-Assessment E-Journal, 1 (2007-14): 1–23.

Comments and Questions

Anonymous - Applied Perspective
January 19, 2008 - 08:18

Literature on structural breaks in panel data is still in the development stage. Unless there is a significant breakthrough many applied economists won't be interested in variations which do not lead to any conclusion. Further, it is also necessary and useful for applied economists if the author's supply the GAUSS ...[more]

... code, data with a useful note on how to replicate. Otherwise you can publish a paper but very few would apply. If that is your main intention that is a different matter.

I suppose you want to improve or make easy Westerlund's method. If so you may use his GAUSS code and supply your version and data. Having said this the choice is yours.

Stefano FACHIN - reply to reader
January 28, 2008 - 14:33

The comment raises two points:
(i) to be of any actual empirical interest, a paper like ours should include GAUSS code;
(ii) what is the value added provided by our paper with respect to Westerlund.
We agree completely with the first point. We simply tought it was unecessary ...[more]

... to state explicitly that our code (indeed written in GAUSS) is available on request, as this is standard routine for published work. Perhaps the journal could request the authors to post the code in a section of the site in order for a paper to be fully accepted, as done by other journals (e.g., the Journal of Applied Econometrics). Although writing an e-mail message is not a great effort, it is likely that having the code readily available for download may indeed encourage more readers to try new procedures.
The second point seems instead to be misplaced. By “Westerlund’s method” the reader probably refers to the panel cointegration test with breaks by Westerlund (2006) ("Testing for Panel Cointegration with Multiple Structural Breaks" Oxford Bulletin of Economics and Statistics 68:101-132). Our paper deals with testing for structural breaks in the cointegrating coefficients given cointegration, a completely different matter.

Stefano FACHIN - Gauss Program
March 31, 2008 - 15:13

A Gauss program suitable for empirical applications is downloadable following this link.