Sunday, December 14, 2008

The calculations that lead up to fraud

Under normal circumstances, the checks and balances of capitalism work fairly well, particularly in good countries, when it comes to the problems of fraud. This reflects the rational decisions of individuals who compare the benefits from two paths:

Behaviour within the rules
This yields the NPV of cashflow from now until death from being able to work and earn profits in the business.

Breaking the rules
This yields some benefits immediately. There is a certain probability of getting caught and a certain delay in getting caught. Once the person is caught, a punishment is inflicted, and the NPV of cashflow from good behaviour is lost.

The system of checks and balances has been optimised for normal times and, by and large, in good countries, it does a good job of deterring misbehaviour.

Understanding the two big recent blowups

The global economic turbulence changes the balance between these elements. For many people who have their backs against the wall, when faced with imminent disaster in their ordinary business, the payoff to the first option -- behaviour within the rules -- goes down sharply. This increases the temptation of breaking the rules.

The Madoff story will inspire some to say that the concept of a hedge fund is fundamentally broken. They will argue that in good times, the checks and balances work out okay, but when volatility is high and some hedge funds have made very large losses, the temptation towards malpractice becomes irresistible, and the lack of hands-on government involvement in hedge funds is a fatal flaw.

The story is a little more complex. A more careful examination shows that both cases (Dreier and Madoff) were a bit out of the ordinary. In the Dreier case, as the NYT article by Alison Leigh Cowan, Charles V. Bagli and William K. Rashbaum says:

Mr. Dreier, 58, controlled the finances of his law firm to an unusual degree, according to lawyers there, because of the unusual way it was set up.
Mr. Dreier was the only equity partner in the firm, and deals were structured so that only he knew all the specifics and had access to all accounts, people with the firm said in court papers. Mr. Dreier convinced lawyers that such an arrangement was best by emphasizing that it would allow them to concentrate on their first love, the law, while he worried about running the firm.
There would be no executive committee. No partners meetings. Mr. Dreier would handle all administrative chores.

Their checks and balances were unusually weak for this organisation, even by the standards of tranquil times.

In Madoff's case, as Roger Ehrenberg says:

Hedge funds, the purported touchstone of the unregulated entity, are far more regulated and subject to many more checks and balances than Madoff every was. I've long made the argument that hedge funds are actually heavily regulated, not directly but indirectly through their relationships with the heavily regulated prime brokers. Forget about the negative PR and spin - it's true. Prime brokers have full transparency into the books of hedge funds, contribute data to the reporting of Net Asset Value (NAV), which is generally pumped out by the hedge funds' administrator. There is a further layer of protection offered by the hedge fund's auditor. Unless everyone is in cahoots it is pretty hard to see how a hedge fund is systematically mis-reporting NAV (except with respect to illiquid assets, but this is another issue entirely).
Some of the biggest non-market risks of hedge funds include style drift (veering from the strategy outlined in the prospectus, such as when Amaranth's natural gas trades ceased to make it a multi-strategy fund), creeping illiquidy (taking advantage of the illiquid asset carve-out in the prospectus only to see the value of the liquid assets fall, resulting in a prospectus-breaching concentration in illiquids), overuse of side pockets (concentrated, balky public positions that don't fall under the rubric of illiquids yet result in a similar risk profile) and manager fatigue ("If I'm down 50% and it will take me years to dig out from under my high water mark, I'll just shut down"). Note that these risks have to do with the character of the manager, things that a good due diligence process should ferret out. But they really don't have to do with the veracity of the firm's positions, books and records, as third-party involvement together with the regulatory oversight of the prime brokers makes the Madoff kind of fraud highly unlikely.
But Madoff is a completely different kind of firm. It is a broker/dealer with an asset management division, enabling it to rely entirely on itself for trading and settlement. Further, it used a no-name, three-person accounting firm, unheard of for a firm of Madoff's size, scope and complexity. A purely rational trader of Madoff's stature would have set up a hedge fund business to extract 2/20 from his clients. I guess we now understand why; it would have subjected his portfolio to the unwanted scrutiny of his prime brokers. By keeping his game completely in-house and on the down low, it essentially fell through the cracks of our regulatory structure. Will this cause the SEC to redouble its efforts in regulating broker/dealers? Force changes in transparency, similar to what I've pushed for in the OTC derivatives market to the broker/dealer community? Or is it simply a matter of creating rules that ensure credible third-party involvement in the validation of assets under management/NAV in order that Madoff's brand self-dealing couldn't be sustained?

When it comes to client funds, I believe the involvement of multiple third-parties in the validation of positions and NAV is critical. Checks and balances have to be built into the system, and by employing a structural approach to regulation as opposed to simply adding more regulations, I believe we can minimize the friction in the system while providing the necessary protections to individuals and institutions. The lack of trust so pervasive in today's financial markets just took another hit. But let's take a moment to think of the right way to address the issue (better due diligence, higher standards for fiduciaries, imposition of checks and balances with broker/dealers and asset managers working under the same roof), rather than the way that plays best for PR purposes.

In this case also, the checks and balances that prevailed on Madoff's operation were not typical of what is found with the ordinary hedge fund. The Madoff story is not an indictment of the concept of a hedge fund, but a critique of the specific form through which his fund was organised.

Benefits from the involvement of a third party

I see an analogy with the idea of the third-party repo.

A repo is a collateralised loan. You give me a security worth Rs.100 and I give you a loan of Rs.80. As the price of the security fluctuates, marking to market needs to be done to ensure safety. The difference (Rs.20) is called the haircut'. The size of the haircut is chosen based on the price risk and liquidity risk of the security between two consecutive marks-to-market.

It is possible to organise a repo properly with bilateral credit risk exposures. If both parties were good and efficient, then marking to market of collateral values would take place properly, haircuts would be computed correctly and always maintained. But there is the risk of operational failures or outright fraud. This is where the third-party repo' greatly improves matters. The borrower and lender do not directly deal with each other: each deals with the 3rd party who supervises the proper functioning of the transaction as time passes.

This helps simply by bringing in a third party into the transaction. It increases the number of people reconciling accounts. But at the same time, if the third party is just an accountant, there is a greater risk of his doing work only on a best efforts basis. What will really bind the incentives of the third party is for him to do netting by novation: for him to be the legal counterparty to both sides. So when X borrows from Y, at a legal level, X borrows from the third-party and the third-party borrows from Y. This ensures incentive compatibility for the third-party who is then much less likely to make mistakes.

An Indian perspective

In India, under normal conditions, blocking fraud is difficult because the probability of being caught is low, the delay in imposing punishment is high, and the punishment is often quite small.
In this backdrop, the events of 2008 have induced massive profits and losses in unexpected places. I suspect some scandals will pop up.

By and large, the world of SEBI, NSE, NSCC, CCIL, BSE, NSDL, CDSL and mutual funds works fairly well in having strong checks and balances. The life of a typical securities firm in connection with these elements of securities infrastructure is tightly integrated into IT systems run from above. These IT systems correspond to a real-time offsite supervision system. They substantially remove room for fraud. While I expect things will work out okay there, there is always the possibility of some chinks in the armour showing up.

The famous scandals of the past are instructive. Harshad Mehta exploited the flaws of the depository for government bonds. Ketan Parekh exploited the weak risk management of Calcutta Stock Exchange. Home Trade exploited the flaws of the settlement system for bonds. Each of these took place in a part of the system where the real-time offsite supervision system described above was absent. That's a fair guide to what might happen in 2009.

Problems are perhaps more likely in the less regulated companies including listed companies. Some firms have a lot of leverage on their balance sheets and some CEOs have a lot of leverage on their personal balance sheets. Some CEOs have personally given buyback promises to institutional investors who got invested in their stock. These individuals are under a lot of pressure. The less ethical of them could buckle under this pressure and resort to breaking the law.

1. your prognosis has come true. SATYAM.

2. See my post on Satyam.

3. This comment has been removed by a blog administrator.

4. Anonymous,

Your comment is very interesting but has some legal ramifications which I am looking into.

5. This comment has been removed by a blog administrator.

6. Ajay, your post is so right. Look at the processions of scoundrels who have been outed after Lehman crashed and the tide flowed out.

Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticism is perfectly okay; uncivilised language is not. We delete any comment which is spam, has personal attacks against anyone, or uses foul language. We delete any comment which does not contribute to the intellectual discussion about the blog article in question.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.