How Much Should Financial Industry Leaders Trust in Academic Research?
Given that governments spend huge sums of money on education and university research, one wonders what is the payoff for business leaders? Take the example of university research in finance and the financial industry. Of course, the empirical results published in financial agencies are relevant for the sector, in particular given the dependence of hedge funds on the investment strategies proposed by researchers in the financial literature. But in this regard, a relevant question is: what is the competence of these researchers, especially those who publish their research results in leading financial journals? Especially if these researchers have never really worked in the sector.
As in any other discipline in the social sciences, finance researchers strive to have their studies published in what, in academic circles, are generally referred to as “top-notch journals” such as the Finance Review, Review of financial studies, Where Financial economics journal. Obviously, in addition to what is commonly associated with recognition among fellow researchers, publication in these journals has a positive impact on a researcher’s academic career.
However, it can be difficult for academics outside of the inner circle to get their studies published in these leading journals, a phenomenon often referred to as “selection bias”. So the crucial question is: Are research studies published in leading financial journals more reliable?
Ironically, recent research on replication failures has shown that articles published in these leading journals are cited, on average, more than 100 times more often than follow-up articles which, however, stand the test of time, but are published in less recognized journals. While previous research argued that ‘picking’ manifested in spec search or p-hacking could be a reason for the high rate of replication failures, a recent Finnish study to study conducted at the University of Vaasa takes a new perspective by exploring whether the nature of financial markets allows us to use standard statistical tools often used in financial research.
The to study, directed by Dr Klaus Grobys, assistant professor at the University of Vaasa and member of the Innovation and entrepreneurship InnoLab, a multidisciplinary, phenomenon-based open research platform, provides strong evidence that standard methods often used in financial research inevitably yield sample-specific results. Grobys argues that traditional methodologies often used in financial research are based on axioms derived from statistics.
These axioms are valid in specific random environments where the past tells us something about the future. These random environments bend the laws of physics. Man-made systems don’t work the same way. Put simply, the critical problem with financial market variables is that uncertainty in the future looks different from uncertainty in the past. Think about the stock market crash of October 1987, which brought about permanent changes in the valuation of stock options. âBecause future uncertainty is different from past uncertainty, standard statistical methodologies fail to provide accurate results in such random environments,â he says.
Studies in leading journals use the same statistical methods, which manifests itself in many replication failures. This casts serious doubts on the quality signal associated with these reviews. The correct use of incorrect methods will not produce any of the necessary advances in this area of ââknowledge. We can ask ourselves: what is the consequence for the financial industry of relying on these methods? Referring to the bankruptcy of Long-Term Capital Management (LTCM) hedge funds, of which Robert Merton Jr. and Myron Scholes were founding partners, best-selling author Nassim Taleb points out that the consequences of relying on bad practices can be destructive. In his book The black swan, he underlines on page 288: ââ¦ during the summer of 1998, a combination of large-scale events, triggered by a Russian financial crisis, unfolded outside their models.
It was a black swan. LTCM went bankrupt and almost destroyed the entire financial system with it, because the exposures were massive. ” In his job The black swan, which has garnered tremendous attention both inside and outside academia, Taleb points to LTCM’s bankruptcy because (I) Merton and Scholes were academics with first-rate resumes from the highest ranks in academia, and (I), the scale of the losses was spectacular. Investors relying on the skill of leading scientists have lost an astonishing amount of their wealth, so it can be argued that this is a serious issue that needs more attention.