Les Oxley

Cointegration, causality and export-led growth in Portugal, 1865–1985

Abstract This paper considers the relationship between cointegration and causality and uses tests of cointegration as pre-tests for Granger tests of causality. Such methods are utilized on Portuguese data, 1865–1985, where the export-led growth hypothesis is rejected in favour of reverse causality.

Linear Saddlepoint Dynamics 'on their Head'. The Scientific Content of the New Orthodoxy in Macrodynamics

Abstract The paper proposes a particular property, robustness, which should be considered a necessary feature of scientifically valid models in economics. The scientific status of the New Orthodoxy in macroeconomics is then examined in the light of this property. It is shown that, in New Orthodoxy macrodynamics, the rational expectations hypothesis (REH), has to be augmented with ancillary hypotheses in order to achieve robustness. These hypotheses are far stronger and less plausible than the REH itself. The article concludes that most of the New Orthodoxy macrodynamics can be rejected on methodological grounds alone.