The recent collapse of Refco, a large US-based broker in the commodities and derivatives market, has several interesting features. (Incidentally, Refco has a subsidiary in India as part of its operations in 14 different countries).

For one, the sheer speed with which developments took place. The second is the involvement of a hitherto obscure Austrian bank. The third, and the most puzzling to my mind, is the questions the case raises for students of behavioural finance.

What happened can be quickly summarised. Refco had grown rapidly (2,500 employees, 650 million derivatives contracts brokered in 2004/05), not only in its tightly regulated broking business, but also in unregulated activities like borrowing and lending of securities.

In the latter segment, it was particularly active with hedge fund clients. Early last month, an employee came across a loan transaction with a hedge fund, which aroused his suspicions. He brought it to the notice of the board. Within days, the board suspended the CEO for fraudulently hiding loans from Refco to private companies controlled by him - the amount was as high as $430 million. On his suspension, the share price collapsed - Refco had gone in for an initial public offering just two months before the fraud came to light.

Amazingly, given the relatively simple way in which the transactions were hidden, they had not been noticed during the course of due diligence by investment bankers and audit by Grant Thompson.

Equally amazingly, within a few days of his suspension, the CEO managed to borrow enough money from an Austrian, trade union-owned bank, to pay off the loan. Incidentally, insiders including Refco's executives and a private equity fund, took away more than a billion dollars from the company in August, out of the sale of shares in the IPO.

It seems that the loan had been on the books of Refco for a couple of years and a simple method was adopted to hide it from auditors and from disclosure requirements in quarterly accounts.

A few days before the end of the quarter, the loan would be transferred to another borrower, a hedge fund in this case, to avoid its disclosure as a related party transaction.

The entry would be reversed within a few days in the new quarter and obviously the hedge fund earned a good fee for, in effect, renting its balance sheet to park a loan for a coupe of weeks. (Indeed, it was the interest rate on that loan which first attracted the attention of the employee).

What the fourth largest Austrian bank, a minor shareholder in Refco, was doing in bailing out the suspended CEO, that too by lending him almost half a billion dollars, is another curious aspect of the case. (It, however, illustrates the globalisation of modern finance: some of the other shareholders of Refco included a Moscow-based hedge fund, firms in Italy, Venezuela and the Bahamas, besides company executives and the US-based private equity fund.)

Since the bulk of the security for the Austrian bank's loan to the Refco CEO consisted of shares in the company itself, it obviously is facing a big loss. The CEO is under house arrest and will be prosecuted for securities fraud.

Refco's unregulated securities business has been closed down, and the regulated brokerage business will be auctioned by the bankruptcy court in a couple of days. From collapse to sell and resumed functioning within one month -- surely our financial, investigative and legal systems need to learn from the episode.

The case also has some very interesting features for students of behavioural finance. In parallel with Enron, WorldCom and umpteen other companies, it is the top executives, whose compensations run into tens of millions of dollars, who seem to be perpetrating the biggest frauds.

What is the motivation? Surely, by almost any yardstick, the fraudulent behaviour should be considered as irrational? Do the risks taken justify the rewards, which, in any case, are not required for any conceivable need of the perpetrator? While one can surely understand why hundreds of people looted shops during the flooding of New Orleans, it is much more difficult to make sense of the misdeeds of top executives.

And, if man is not rational, what happens to the massive body of economic theory based on the assumption of rational behaviour?

Tailpiece: Only Germany and Austria seem to have traditions of trade unions owning commercial banks. I recall that George Fernandes took inspiration from the German trade union bank to promote the Bombay Labour Cooperative Bank a long time back, to lend money to taxi and rickshaw drivers whom commercial banks were not willing to finance. Incidentally, the German trade union bank is also in deep trouble after mismanaging the interest rate risk on the gaps in its assets and liabilities.