Prof. Jayanth R. Varma's Financial Markets Blog

Photograph About
Prof. Jayanth R. Varma's Financial Markets Blog, A Blog on Financial Markets and Their Regulation

© Prof. Jayanth R. Varma

Subscribe to a feed
RSS Feed
Atom Feed
RSS Feed (Comments)

Follow on:

Sun Mon Tue Wed Thu Fri Sat

Powered by Blosxom

Mon, 28 Mar 2016

Regulatory priority: punish, deter or protect?

When a serious breach of market integrity is suspected, what should the regulators’ priorities be: should it try to punish the guilty, or should it seek to deter other wrong doers or should it focus on protecting the victims? Both bureaucratic and political incentives may be tilted towards the first and perhaps the second, but in fact it is the last that is most important. I have been thinking about these issues in the context of the order of the Securities and Exchange Board of India in the matter of Sharepro Services, a Registrar and Share Transfer Agent regulated by SEBI. The order which is based on six months of investigation and runs into 98 pages finds that:

If one assumes that these findings are correct, then the key regulatory priority must be to take operational control of Sharepro and thereby protect the interests of investors who might have been harmed. A Registrar and Share Transfer Agent is a critical intermediary whose honest functioning is essential to ensure market integrity and maintain the faith of investors in the capital markets. I think that SEBI’s powers under section 11B of the SEBI Act would be adequate to achieve this objective, but in case of need, resort could also be had to section 242 of the Companies Act 2013.

The SEBI order does take some steps to punish the top management of Sharepro but does too little to protect the investors who appear to have lost money. It does not even cancel or suspend the registration of Sharepro as a Registrar and Share Transfer Agent, but merely advises companies who are clients of Sharepro switchover to another Registrar and Share Transfer Agent or to carry out these activities in-house. The only investor protection step in the order is the direction to companies who are clients of Sharepro to audit the records and systems of Sharepro. But if the records have been falsified, then only a regulator or other agency with statutory powers can carry out a meaningful audit by obtaining third party records.

A decade ago, when the Satyam fraud occurred, I was among the earliest to write that the government should simply take control of the company. I would argue the same in the case of Sharepro as well assuming that the SEBI findings are correct.

Posted at 19:01 on Mon, 28 Mar 2016     View/Post Comments (5)     permanent link

Sun, 20 Mar 2016

Why would a manufacturer finance a bank's capital expenditure?

Usually a manufacturing company goes to a bank to finance its capital expenditure. But last month witnessed a deal where the US manufacturing giant GE stepped forward to finance the capital expenditure of one of the largest banks in the world – JPMorgan Chase – when the latter decided to buy 1.4 million LED bulbs to replace the lighting across 5,000 branches in the world’s largest single-order LED installation to date.

As a finance person, the first explanation that I looked at was the credit rating. GE lost its much vaunted AAA rating during the Global Financial crisis, but based on S&P long term unsecured ratings, GE’s AA+ rating is full five notches above JPMorgan Chase’ A- rating. Based on Moodys ratings, the gap is only two notches. Averaging the two and taking into account S&P’s negative outlook on GE, we could say that GE enjoys a rating that is a full letter grade (three notches) above JPMorgan. So perhaps, it makes sense for the manufacturer to finance the bank.

Another possible explanation is that trade credit has a set of advantages that are not fully understood. Some of the alleged advantages of trade credit (like the idea that a business relationship leads to superior information on credit worthiness) strain credulity when the recipient of the credit is one of the largest banks in the world with hundreds of publicly traded bonds outstanding. Similarly, the idea that vendor financing is a superior form of performance guarantee is hard to believe when the vendor is a manufacturing giant with such a high reputation and credit rating.

In a Slate story, Daniel Gross explained the logic in terms of the inefficiency and inertia of corporate bureaucracies:

The second barrier – “the capital barrier,” as Irick call it – is more difficult to surmount. The economics of buying and installing them can be a challenge to corporate bureaucrats. Companies often produce multiyear budgets well in advance. Going LED means spending a lot of money in a single year to buy and install them, make sure they work, and dispose of the old ones. And it is difficult even for a company like Chase to make a decision quickly to write a check to buy 1.4 million new light bulbs and pay for their installation. GE, of course, has a long track record of helping to finance customers’ purchases of its capital goods, structuring payments over a period of years rather than upfront.

That perhaps makes more sense than any of the finance theory arguments.

Posted at 11:14 on Sun, 20 Mar 2016     View/Post Comments (0)     permanent link

Fri, 18 Mar 2016

Random Justice

Voltaire wrote that “His Sacred Majesty Chance Decides Everything”. Neustadter comes to a similar conclusion in a fascinating paper entitled “Randomly Distributed Trial Court Justice: A Case Study and Siren from the Consumer Bankruptcy World” (h/t Credit Slips):

Between February 24, 2010 and April 23, 2012, Heritage Pacific Financial, L.L.C. (‘Heritage’), a debt buyer, mass produced and filed 218 essentially identical adversary proceedings in California bankruptcy courts against makers of promissory notes who had filed Chapter 7 or Chapter 13 bankruptcy petitions. Each complaint alleged Heritage’s acquisition of the notes in the secondary market and alleged the outstanding obligations on the notes to be nondischargeable under the Bankruptcy Code’s fraud exception to the bankruptcy discharge. ...

Because the proceedings were essentially identical, they offer a rare laboratory for testing the extent to which our entry-level justice system measures up to our aspirations for ‘Equal Justice Under Law.’ ...

The results in the Heritage adversary proceedings evidence a stunning and unacceptable level of randomly distributed justice at the trial court level, generated as much by the idiosyncratic behaviors of judges, lawyers, and parties as by even handed application of law ...

Neustadter summarizes the outcome of these proceedings as follows (Table 1, page 20):

Recovery by Heritage Filed Settlement Agreements Heritage Requests Dismissal Dismissal for Other Reasons Default Judgments Summary Judgments Trials
Zero 49 26 12 N/A 4 3
Positive 103 ($1m) N/A N/A 10 ($0.9m) 1 ($0.06m) 2 ($0.2m)

I remember reading Max Weber’s Economy and Society decades ago and being fascinated by his argument that legal rights only increase the probability of certain outcomes (incidentally, Weber obtained a doctorate in law before becoming an economist and sociologist). Weber believed that the function of law in a modern economy was to make things more predictable, but by this also he only meant that probabilities could be attached to outcomes. I resisted Weber’s argument at that time, but over the course of time, I have come around to accepting them. In fact, I now think that it is only the conceit of false knowledge that leads to a belief that certainty is possible.

A greater degree of acceptance of randomness would make litigation a lot more efficient. In my view of things, a judge should be required to set a time limit for the amount of time to be devoted to a particular dispute (depending on the importance of the dispute). When that time has been spent, the judge should be able to say that he thinks there is say a 40% probability that the plaintiff is right and a 60% chance that the defendant is right. He should then draw a random number between 0 and 1; if the number that is drawn is less than 0.4, he should rule for the plaintiff, otherwise for the defendant. All litigation could be resolved in a time bound manner by this method. Even greater efficiency is possible by the use of the concepts of Expected Value of Perfect Information and Expected Value of Sample Information to decide when to terminate the hearings and proceed to drawing the random numbers. If an appeal process is desired, then of course the draw of the random number could be postponed until the appeal process is exhausted and the final value of the probability determined. In a blog post a couple of months ago, I have discussed cryptographic techniques to draw the random number in a completely transparent and non manipulable manner.

In my experience, there is enormous resistance to deciding anything by a draw of lots or other randomization technique though I believe that it is the most rational way of decision making. Instead society creates very complex mechanisms that lead effectively to a process of randomization based on which judge gets to hear the matter and what procedural or substantive legal provisions the judge or the lawyer is aware of. In fact, one way of making sense of the bewildering complexity of modern law is that it is just a very costly way of achieving randomization – if the law is too complex to be remembered by any individual, then what provision is remembered and applied is a matter of chance. That is how I interpret Neustadter’s findings.

In case you are wondering why I am discussing all this in a finance blog, let me remind you that the litigation in question was about recovery of defaulted debt and that is definitely a finance topic.

Posted at 14:50 on Fri, 18 Mar 2016     View/Post Comments (1)     permanent link

Sat, 12 Mar 2016

Bangladesh Bank hacking is yet another wake up call

A year ago, I blogged about the Carbanak hacking and thought that it was a wake up call for financial organizations to improve their internal systems and processes to protect themselves from patient hackers. The alleged patient hacking reported this week at the central bank of Bangladesh shows that the lessons have not been learned. There is too much of silo thinking in large organizations – cyber security is still thought to be the responsibility of some computer professionals. The reality is that security has to be designed into all systems and processes in the entire organization. Institutions like central banks that control vast amounts of money need to defend in depth at all levels of the organization. Physical security, hardware security, software security and robust internal systems and processes all contribute to a culture of security in the whole organization. In my experience, even senior management at large banking and financial organizations have a highly complacent attitude towards security that makes the organization highly vulnerable to a patient and determined hacker.

For example, there is no reason not to have a dedicated terminal for large (say $100 million) SWIFT transactions. Cues like dedicated hardware tends to make humans more alert to security considerations. In the paper world, we went to great lengths to institutionalize such cues. For example, the law on cheques permits cheques to be written on plain paper (the law only says “instrument in writing”), but in practice it was always written on special security paper. The importance of keeping blank security paper under lock and key was drilled into every person who worked in a bank from the chairman to the messenger boy. I have yet to see any similar attempt to inculcate a culture of computer security in any bank.

Posted at 22:03 on Sat, 12 Mar 2016     View/Post Comments (0)     permanent link

Sun, 06 Mar 2016

Investment Banks and IPO Pricing Power

Krigman and Wendy have an interesting paper on how issuers pay for their investment banks’ past mistakes. Their conclusions are based on the IPOs that came to market after the the botched Facebook IPO of 2012 in which the stock fell below the IPO price and the investment banks had to buy shares in the market to stabilize the price. IPOs after this event were underpriced by an average of 20% compared to only 11% prior to the Facebook IPO. More interestingly:

We show that the entire increase in underpricing is concentrated in the IPOs of the Facebook lead underwriters. We find no statistical difference in underpricing pre and post-Facebook for non-Facebook underwriters. We argue that investment bank loyalty to their institutional investor client based propelled the Facebook underwriters to increase underpricing to compensate for the perceived losses on Facebook.

“Loyalty to investor client” sounds very nice in a scandal dominated era where we have to come to believe that bankers have no loyalty to anybody. Yet, it must be remembered that the alleged generosity to investor clients did not come out of the bankers’ profits; it came out of the pockets of another bunch of clients – the issuers. This raises a very disturbing question: what gives them the pricing power to underprice issues relative to what their competitors were doing? The first possibility that came to my mind is that these were deals that the bankers had already won and it was difficult for the clients to change their lead banks after they had already been chosen. However, the data seem to show that the effect lasted more than a year, and moreover there was a 41 day period following Facebook during which there were no IPOs at all. The other possibility is that this is not a competitive market at all and the investment banks have a lot of market power. Chen and Ritter wrote a famous paper about this at the turn of the century (“The seven percent solution.” The Journal of Finance 55.3 (2000): 1105-1131).

Posted at 16:44 on Sun, 06 Mar 2016     View/Post Comments (0)     permanent link

Wed, 02 Mar 2016

Were most crisis era bank acquisitions failures?

JPMorgan Chairman Jamie Dimon states in a Bloomberg interview that he now regards JPMorgan’s acquisition of Bear Stearns and of Washington Mutual during the global financial crisis as “mistakes”. I used to think that these were among the better deals in the whole lot of crisis era acquisitions which include such monumental disasters as Bank of America’s acquisition of Countrywide or Lloyds’ acquisition of HBOS. But Dimon says that the Bear Stearns purchase ended up costing JPMorgan $20 billion while if I remember right the headline acquisition cost was only a little over $1 billion (and that after Dimon raised the price per share from $2 to $10). If even JPMorgan’s relatively good deals ended up being big mistakes, then I wonder whether the only sound crisis era banking acquisition might be Wells Fargo’s acquisition of Wachovia. Of course, the very best deals were Warren Buffet’s minority stakes in Goldman Sachs and GE, but these do not count as acquisitions. On a purely accounting basis, the US government did make money on many of its rescue deals, but this accounting does not include the hidden costs that contribute heavily to the $20 billion price tag that Dimon now puts on the Bear deal. What all this means is that even at the depths of the global financial crisis, it would have made a lot of sense to heed the good old advice not to try to catch a falling knife.

Posted at 13:30 on Wed, 02 Mar 2016     View/Post Comments (0)     permanent link