When defending one of their many ill-conceived theories, Economists will frequently mention the theories appeared in peer reviewed academic journals.

For those of you who don’t know, the peer review process is suppose to be the gold standard as it is suppose to vet the paper in which the theory appears prior to publication.  As a result, even an ill-conceived, obviously flawed theory that is published is instantly rendered credible.

Unfortunately, the peer review process has one glaring weakness that lets these ill-conceived, obviously flawed theories through:  the reviewers.  By definition, the quality of the review is only as good as the reviewers.  If the reviewers aren’t any good, the peer review process isn’t any good.

Let me give you an example I have written about many times:  Gary Gorton and Informationally Insensitive Debt.  He offers as examples of this type of debt: deposits in banks and the AAA-rated tranches of structured finance securities.

As soon as the reviewers saw this, they should have rejected it.

Why?

Because these are examples of informationally sensitive debt.  This isn’t surprising since our financial system is designed so all debt is informationally sensitive.  Why?  Investors know the tradeoff for being provided with disclosure so they can know what they own is they are responsible for all losses on their investment exposures.  Hence, investors are always looking for new disclosures.

But why didn’t the editors and reviewers know this and reject the article?

Take deposits in a bank.  The information deposits are sensitive to is the amount covered by the government deposit guarantee.  This should have taken the reviewers a nano-second to realize.

Take AAA-rated tranches of structured finance securities.  These tranches are also sensitive to information.  Specifically, to information on the current performance of the underlying collateral.  This might have taken a whole minute of thinking to realize.

What Professor Gorton identified as informationally insensitive was simply how easy or hard it was to use the necessary information for assessing the risk of the debt.

The fact his paper was published in a leading peer-reviewed Economic journal tells you just how low the standard is for peer-review in Economics.  I would hazard to say the bar an Economics article has to get over to get published is buried several feet under ground.

After all, how else can you account for the apparent lack of training the reviewers received in understanding how our financial system is designed?

---

It is clear this post offended some in the Economics profession.  If you found it offensive, it is up to you to fix the peer review process.  You cannot but the burden on people outside of the Economics profession to show how articles like Professor Gorton’s are fundamentally flawed.

When you let articles like Professor Gorton’s get published, it undermines the credibility of the articles where the peer review process worked as it should.  After all, unless the reader is an expert, they have no way of telling if the article was a peer review success or a peer review failure.  However, once readers become aware of the prevalence of peer review failures in Economic journals, even peer review successes are tainted.

---

I received pushback that I cherry-picked the Gorton article because it was so easy to debunk.  Actually, I could have chosen hundreds of articles that have been published in Economic journals that find a flaw with transparency.

A classic argument is the idea too much transparency is as bad as too little transparency as it confuses the investor.

Again, a nano-second of thought by the editors and reviewers should tell them this is wrong.

First, there is no law that says an investor must do the assessment of the risk of an investment themselves.  They are free to hire trusted third party experts to do the assessment for them.  Proof investors do this can be seen in the explosive growth of the asset management industry.

Second, too much transparency doesn’t exist.  What if Bank of America disclosed all of its current exposure details?  This would be a lot of disclosure and at first blush appears to fit the definition of “too much” transparency.  But is it?  Consider that JP Morgan is an investor through the myriad exposures it has in Bank of America.  Do you think JP Morgan could assess this disclosure?  If you don’t, the problem isn’t a flaw with the disclosure, but rather a feature.  If Bank of America is too big for JP Morgan with all of its analytical resources (people and computer systems) to assess, there is no reason to think financial regulators or Bank of America’s management can.  Therefore, it needs to be shrunk until it can be assessed.

---

Wow, that pushback on my debunking of peer review of the too much transparency literature was fast!

Please note, despite its existence for a decade, I am not holding the Economics profession responsible for knowing or understanding the Information Matrix and therefore reviewing a significant amount of the Economic literature within this framework.

Rather, I am holding the members of the Economics profession responsible for not reviewing a paper where you either don’t have the expertise to evaluate the contents or are unwilling to spend a minute critically examining the underlying assumptions the authors make.

I know when it comes to transparency members of the Economics profession have found adhering to this standard very difficult.  Most see themselves as experts due to the vast literature on Perfect Information and Information Asymmetry.

Unfortunately, this literature has a huge hole in it.  This hole was filled by the Information Matrix and the formal recognition a significant percentage of the transactions in the financial marketplace don’t occur where there is Perfect Information or Information Asymmetry.  Rather, they are simply blind bets where the underlying security doesn’t provide the disclosure necessary for either buyer or seller to know what they own.  Why do these trades occur?  Think Wall Street phishing for fools and behavioral economics.