Academic journal article Canadian Journal of Administrative Sciences

Cointegration and Canada-U.S. Team Structures: Does Accounting for Structural Breaks Make a Difference?

Academic journal article Canadian Journal of Administrative Sciences

Cointegration and Canada-U.S. Team Structures: Does Accounting for Structural Breaks Make a Difference?

Article excerpt

Abstract

Using weekly data on Canadian and U.S. T-bill and Tbond yields over the 1982 to 1996 period, we find evidence in support of the generalized interest parity model (i.e., co-movement) for T-bond term structures, but not for T-bill term structures. This evidence is robust to the introduction of fractional cointegration. However, structural breaks induced by political uncertainty surrounding Quebec separation and Canada's move to zero inflation had larger impacts on the T-bill term structure than on the T-bond term structure. This paper highlights the importance of accounting for structural breaks, and how a failure to do so may affect statistical analyses undertaken by both financial economists and practitioners. As an example, we discuss the implications for portfolio management. In particular, incorrect conclusions that Canadian and U.S. T-bill markets are not cointegrated can result in a less than optimal amount of diversification, thus affecting the performance of portfolio managers.

Resume

En utilisant des donnees hebdomadaires de rendements sur les effets et bons du Tresor canadiens et americains pendant la periode de 1982 a 1996, nos resultats soutiennent le modele generalise de la parite des taux d'interets (co-mouvement) dans le cas des structures des echiances des bons du Tresor et non dans celui des effets du Tresor Cette evidence est robuste a l'introduction de la co-integration fractionnelle. Cependant, les ruptures structurelles introduites par l'incertitude politique que presentent la separation du Quebec et les politiques anti-inflationaires canadiennes ont eu un impact plus important sur les structures des icheances des effets que sur ceux des bons du Tresor, ce qui parvient a expliquer les refus de la theorie. La presente etude souligne Vimportance de l'incorporation des ruptures structurelles sans laquelle les analyses statistiques entreprises par les economistes de finance et par les practiciens seraient inversement affectees. Entre autre, nous presentons les implications qu'ont les ruptures structurelles sur la gestion de portefeuilles. En particulier, la conclusion incorrecte que les marches des effets du Tresor canadiens et americains ne sont pas sont cointegres (lorsqu'ils le sont) menerait a un niveau de diversification sousoptimal, et donc, affecterait negativement la performance des gestionnaires de portefeuille.

Canada's move to zero inflation in February 1988 was a significant policy shift on the part of the Bank of Canada. A similar policy was not adopted in the U.S. This change in relative inflation policies of the two countries has had important implications for the Canadian term structure of interest rates. In addition, uncertainty surrounding the future of Quebec which manifested itself in two referendums on sovereignty and one on the Charlottetown Accord, has had implications for the Canadian term structure. This paper highlights the importance of accounting for such "structural breaks" in order to better understand movements in the Canadian term structure as well as its co-movement with the U.S. term structure.

In an open economy, international arbitrage results in a link between yields on domestic and foreign bonds with the same maturity. In particular, if one assumes that changes in exchange rates and country-specific risk premia are stationary, then Canadian and U.S. yields on bonds with the same maturity should be cointegrated. That is, these yields should exhibit co-movement. As pointed out by Boothe (1991), cointegration is consistent with the hypothesis of high or perfect substitutability between Canadian and U.S. securities. Furthermore, this would imply that Canadian and U.S. securities share the same set of fundamentals and hence the same set of systematic risks. On the other hand, a lack of cointegration would indicate that a different set of fundamentals drives markets in each economy, thus implying important diversification benefits to investing in both markets. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.