When the acute phase of the Great Financial Crisis hit, PhD Economists went into overdrive inventing new theories to explain what happened.  Their theories reminded me of throwing jello against a wall to see what would stick.  Remarkably, one theory actually stuck.

The theory that stuck described financial panics as the result of debt that was previously information insensitive suddenly becoming information sensitive.  MIT’s Nobel Prize winning Economist Bengt Holmstrom gave a speech at a 2014 BIS conference explaining the Information Insensitive Debt theory.

The speech is remarkable for many reasons.  The foremost reason being the audience apparently had and still has no idea why the theory is wrong or how, if it is right, it undermines the vast majority of Economic models and all of Finance.

Professor Holmstrom used the following chart to illustrate this theory.

Please pay special attention to the Information Sensitive region and the Information Insensitive region.

For his analysis, Professor Holmstrom uses money market debt that is backed by some type of collateral.  When the value of the collateral is greater than the debt payoff, the value of the debt is in the Information Insensitive range.  In this range, he asserts investors don’t care about information.  When the value of the collateral approaches or is less than the debt payoff, the value of the debt is in the Information Sensitive range.  In this range, he claims investors care about information.

So what flips the switch so investors go from not caring to caring about information about the value of the collateral?  It certainly cannot be any information about the value of the collateral because that would imply investors are never information insensitive.

He doesn’t say.

Perhaps under the Information Insensitive Debt theory the information sensitivity switch is flipped by magic.

Let me offer an alternative theory:  investors are always sensitive to information affecting the value of the collateral backing the debt.  This information sensitivity starts from the moment investors begin considering an investment in the debt and lasts until the principal and interest payments at maturity.

Why?

The investors know the borrower might default on the debt.  If this happens, the investors will be left with the collateral.  Collateral they will have to sell in the hopes of getting their investment back plus any promised returns.  Hence, investors are always sensitive to information affecting the value of the collateral.

I can understand why Professor Holmstrom’s chart might confuse economists (particularly someone like Ben Bernanke who was desperate for an explanation of the behavior of the opaque debt that fit his financial crisis narrative).  If you take away the references to information sensitivity/insensitivity, Professor Holmstrom’s chart is the same chart you would draw up for investors who are always sensitive to information.

The chart Professor Holmstrom should have drawn up to avoid confusion was a step-function chart where the value of the debt drops at the transition point from information insensitive to information sensitive.

Now let me focus specifically on Professor Holmstrom’s story about investors and information insensitivity/sensitivity.  He begins by observing

What appears so puzzling in hindsight is that a slew of new complex and opaque products appear to have been so poorly understood even by the experts on Wall Street. As Michael Lewis wonders in his best-seller study The Big Short: How could Wall Street trade in securities that they knew so little about? Why did no one ask questions?

He suggests that the purpose of opaque securities was to deceive investors. But it is hard to believe that they would be colluding massively to defraud investors.

There is nothing puzzling about this.  In the 1930s the Pecora Commission investigated Wall Street.  It found Wall Street prefers to sell high margin, opaque securities where it can deceive investors over low margin, transparent securities.

Wall Street didn’t have to “collude” to defraud investors.  Every banker and trader understood deceiving investors with opaque securities the investors relied on Wall Street to value resulted in much higher levels of compensation.

Professor Holmstrom goes on to say

I will argue that a state of “No Questions Asked” is the hallmark of money market liquidity; that this is the way money markets are supposed to look when they are functioning well. The near-universal calls for pulling the veil off money market instruments and making them transparent reflect a serious misunderstanding of the logic of debt and the operation of money markets. This misunderstanding seems to be rooted in part in the public’s view that a lack of transparency must mean that some shady deals are being covered up.

If true, this statement undermines most Economic models and all of Finance.

Why?

Most Economic models and all of Finance assume the buyer and seller know what they are buying or selling.  What Professor Holmstrom has just said is blind betting produces the same beneficial outcomes as informed decision making when buyer and seller know what they are buying and selling.

How does he justify his statement transparency isn’t needed?

It is unclear exactly how such transparency is to be achieved for a simple reason: money markets do not have the ability to establish prices for the collateral that backs up the financial claims and it is hard to see how they could. They are bilateral markets, not exchanges. More importantly, I will argue that money markets are expressly designed to obviate the need for price discovery.

Wow!

Professor Holmstrom just said investors are clueless about how to value collateral.

There is absolutely no evidence that investors provided with the information they need to know what they own are unable to use this information to value what they own.  None!

But that doesn’t stop him from making the claim.

This needs to be repeated.  Investors know how to value securities and collateral when they are given the information they need to do this.  This is why our financial system is based on transparency and issuers are required to make disclosures.

Why does this matter? It matters because a wrong diagnosis of a problem is a bad starting point for remedies.

I couldn’t have said it better myself Professor.  Your diagnosis of the problem is wrong and is a really bad starting point.

At this point, Professor Holmstrom is so far down the rabbit hole there is no stopping his championing opacity.  And that is what he does.

People often assume that liquidity requires transparency, but this is a misunderstanding. What is required for liquidity is symmetric information about the payoff of the security that is being traded. Without symmetric information there is a risk that adverse selection will stick up its ugly head and prevent trade from taking place or in other ways impair the market….

it is a short step to see that obfuscation may be beneficial, because when no trader knows anything of relevance – when there is “symmetric ignorance” about the payoff – the market will be completely free of fears of adverse selection and therefore very liquid. By contrast, transparency can be harmful if the information released makes private information relevant.

There is no misunderstanding.  Liquidity requires transparency.  Symmetric information about the payoff of the security being traded is not sufficient to guarantee liquidity.

Let me illustrate this using the Information Matrix.

Information Matrix

	Does Seller Know What They Are Selling?
Does Buyer Know What They are Buying?		Yes	No
	Yes	Perfect Information	Antique Dealer Problem
	No	Lemon Problem	Blind Betting

If you were designing a financial system which quadrant of the Information Matrix would you want transactions to occur in?  The only quadrant the Economics profession has shown that delivers positive results is the Perfect Information quadrant.  It is also the quadrant the design of our financial system attempts to make all transactions occur in.

Wall Street knows everyone likes a good story.  The Perfect Information quadrant is the only quadrant where investors can Trust the story they are told and then Verify its accuracy.

Unfortunately, investors don’t always verify Wall Street’s stories.  As a result, transactions occur in the Blind Betting quadrant.  And it is from this quadrant financial crises emerge.  They emerge when the story used to value the opaque security is called into doubt.  When this happens, there is no logical stopping point in the downward valuation of these opaque securities other than zero.  Hence, owners of these securities have an incentive to “run” to try to get their money back as soon as the valuation story is called into doubt.  If the presence of opaque securities is sufficiently large, the result of this run is a financial crisis.

Let us return to the Professor’s observation liquidity requires buyer and seller have identical information about the payoff of the security.  This is true in the Perfect Information quadrant.  Transparency means both buyer and seller have the information they need to know the payoff of the security.  It is impossible to create a financial crisis type run on these securities because buyer and seller can use the disclosed information to determine what the logical stopping point is for the downward valuation of these securities.  As a result, there is liquidity (ie, buyers) for these securities.

His observation is also true of securities in the Blind Betting quadrant.  Buyer and seller have identical information.  Neither buyer or seller knows what the payoff of the security is.  They have what he refers to as “symmetric ignorance”.  Unfortunately, when doubt is cast upon the valuation story told by Wall Street, trading in these securities freezes.

The markets that froze during the Great Financial Crisis were all opaque.  For example, the interbank lending market froze because banks with deposits to lend could not determine if the opaque borrowing bank could repay them.  For example, the market for opaque subprime mortgage-backed securities froze as nobody could value these securities.

Professor Holmstrom digs himself into a deeper hole.

There are several papers that show that one may want to limit publicly available information or hide information to make public information less revealing. For instance, if a bank does not show the composition of its assets in detail then news about the economy may not be as valuable for updating one’s beliefs about the prospects of the bank’s collateral. The bank will appear more opaque which can be valuable….

Banks are opaque because they don’t show the composition of their assets in detail.  As a result, we have bank runs because no one can determine if they are viable or not.

Of course, banks being opaque is valuable to someone.  It is valuable to the bankers.  Opacity allows the bankers to hide the amount of risk they are taking on.  Opacity also allows the bankers to hide their engaging in misbehavior.

Intentional opacity is a rather ubiquitous phenomenon.

The fact intentional opacity is ubiquitous doesn’t make it a good thing.

He cites how De Beers sells diamonds without giving buyers a chance to inspect the packet of diamonds before agreeing to buy them.  This only works so long as the packet of diamonds contains exactly what De Beers represents it contains.  As soon as it doesn’t, De Beers is out of business.  The first buyer who is cheated by De Beers tells all the other buyers to stop doing business with De Beers.

Professor Holmstrom then addresses the ill consequences of debt and opacity.

The occurrence of panics supports the main informational thesis that is being put forward here. Panics happen when information insensitive securities and institutions turn into information sensitive ones as in Figure 1. A regime shift occurs from a state where no one feels the need to ask detailed questions, because it is not worthwhile, to a state where there is enough uncertainty that some investors begin asking questions and finding out information. This can lead to rapid drops in prices and reduced liquidity of the securities that were issued and the banks that underwrote them. Such events can be cataclysmic precisely because the original liquidity of the securities rested on trust rather than a precise evaluation and discovery of collateral values. A panic is an information event that shatters the shared understanding and beliefs on which liquidity rested.

The occurrence of panics doesn’t support his informational thesis.  He still lacks a trigger for investors becoming information sensitive which doesn’t show investors were always information sensitive.  Going from a state of certainty to a state of uncertainty only occurs in the presence of information.

The occurrence of panics supports the thesis put forward under the Information Matrix.  Under the Information Matrix thesis panics happen involving opaque securities and not transparent securities.  That is what even Professor Holmstrom finds.

Finally, he offers up what is the fundamental problem with the Information Insensitive Debt theory.

What one would like to know is whether panics could be eliminated or at least predicted so that measures could be taken to contain them. But we do not know what may trigger a crisis nor do we understand the dynamics of panics well. Precautions can be taken, but the path the panic will take if it starts is as unpredictable as a forest fire or an avalanche. In the wake of the recent financial crisis, there is little reason to believe panics can be eliminated.

The Information Insensitive Debt theory doesn’t shed any light on how to eliminate panics.

Please note, the Information Matrix does shed light on how to eliminate panics.  It describes in detail the measures that should be taken to contain panics as well as how to end them.  The Information Matrix even goes so far as to predict panics will occur where they do.  This is along the opacity fault line in the global financial system.