Article Detail

The Failure of Financial Econometrics: Assessing the Cointegration “Revolution”

Journal 32: Applied Finance

Imad Moosa

One aspect of the failure of financial econometrics is the use of cointegration analysis for financial decision making and policy analysis. This paper demonstrates that the results obtained by using different cointegration tests vary considerably and that they are not robust with respect to model specification. It is also demonstrated that, contrary to what is claimed, cointegration analysis does not allow distinction between spurious relations and genuine ones. Some of the pillars of cointegration analysis are not supported by the results presented in this study. Specifically it is shown that cointegration does not necessarily imply, or is implied by, a valid error correction representation and that causality is not necessarily present in at least one direction. More importantly, however, cointegration analysis does not lead to sound financial decisions, and a better job can be done by using simple correlation analysis.

In the second half of the 1980s, specifically following the publication of the “seminal” paper of Engle and Granger (1987), the world of academic finance and economics experienced a “revolution” similar to that experienced by the worlds of music and dancing as a result of the introduction of rock and roll and the twist. Engle and Granger formalized their work on cointegration and error correction and subsequently adapted the causality test of Granger (1969) to take into account the possibility of cointegration. The introduction of these techniques has created a thriving industry with a rapidly growing output of papers written by academics testing theories in economics and finance that were previously tested using straightforward regression analysis.

Tens of thousands of papers and PhDs later, it is about time to ask whether or not the cointegration “revolution” has changed our lives and led to discoveries that enhance our understanding of the working of the economy and financial markets, which is presumably the objective of scientific research. One would tend to imagine that, since this work was awarded the Nobel Prize, it must be valued the same way as the discovery of penicillin, which was awarded the same prize. However, it seems to me that while cointegration analysis has provided the means for finance and economics academics to get their promotion and students to obtain their PhDs, the technique has contributed almost nothing to the advancement of knowledge.

The objective of this paper is to demonstrate that cointegration analysis, error correction modeling, and causality testing are misleading, confusing, and provide a tool for proving preconceived ideas and beliefs. More important, however, is the hazardous practice of using the results of cointegration analysis to guide policy and financial operations, including investment, financing, and hedging. With the help of examples on stock market integration and international parity conditions it will be demonstrated that cointegration analysis produces results that tell us nothing and that for practical purposes these results are useless at best and dangerous at worst.


Leave a comment

Comments are moderated and will be posted if they are on-topic and not abusive. For more information, please see our Comments FAQ
This question is for testing whether you are a human visitor and to prevent automated spam submissions.