Sunday’s NYT had a great interactive graphics by Amanda Cox detailing the dynamics of recessions and recoveries. One interesting graph pertained to the OECD Leading Indicators:
Using OECD composite leading indicators (CLI), we assess empirically whether the ability of the country-specific CLIs to predict economic activity has diminished in recent years, e.g. due to rapid advances in globalisation. Overall, we find strong evidence that the CLI encompasses very useful information for forecasting industrial production, particularly over horizons of four to eight months ahead. The evidence is particularly strong when taking cointegration relationships into account. At the same time, we find indications that the forecast accuracy has declined over time for several countries. Augmenting the country-specific CLI with a leading indicator of the external environment and employing forecast combination techniques improves the forecast performance for several economies. Over time, the increasing importance of international dependencies is documented by relative performance gains of the extended model for selected countries.
Here’s some relevant graphs from the updated version of the paper (not online):
Chart 3: from Fichter et al. (June 2009 version)The authors note that while there appears to be some degradation over time of the predictive power of the own-country CLI for own-country industrial production, perhaps due to the effect of globalisation — it’s also true
“that performance gains for larger economies (such as Japan and the US) from including the CLI in the analysis appear to be more pronounced and less subject to a deterioration over time. This possibly reflects a lower dependency of these large economies on international business cycles and, hence, a smaller loss of information of the domestic CLI owing to globalisation. In contrast, smaller economies showed a stronger deterioration in the forecasting power of the domestic CLI.”
Those interested in the predictive characteristics of the OECD CLI’s should definitely take a look at this paper.
Originally published at Econbrowser and reproduced here with the author’s permission.