Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Wednesday, November 2, 2011

What went wrong at MF Global

Here is a nice quick summary of what went wrong: CNBC video (short, sweet, and to the point.)

It's the same old song. Big financial firm makes a big bet. Information flows relating to probable payoffs suddenly signal higher risk. Rating agencies move in to downgrade firm's liabilities. A "run" ensues (widespread redemptions) leading to...a liquidity event? or a solvency event? (only time will tell, I suppose.)

Why does MF Global matter? One reason is that the institution was added to the Fed's list of 22 primary dealer banks just 8 months ago. See WSJ article below.

Fed Takes Collateral Damage in MF Global Meltdown
Tue, 1 Nov 2011
698 words
By Min Zeng
November 1, 2011, 10:16 AM ET

The Federal Reserve is among those feeling the pinch from the collapse of MF Global Holdings Ltd., which only eight months ago was added to the Fed’s list of 22 primary dealer banks.

MF Global’s spectacular downfall seems unlikely to pose a systemic financial risk to either the U.S. economy or the Fed, in sharp contrast to the fallout from Lehman Brothers in 2008. But it’s possible it will make the selection procedure tougher for primary dealers, an elite group of institutions with which the New York Fed conducts monetary policy and which are obligated to participate in U.S. Treasury debt auctions.

MF Global’s fortunes quickly went downhill over the past week amid concerns over its exposure to the euro zone’s sovereign debt. In this way, its travails underscore the potential contagion risks to the U.S. financial system via the primary dealer network. Besides MF Global, several primary dealers are owned by big European banks, including France’s BNP Paribas SA and Societe Generale SA, whose shares sold off in September due to concerns about their exposure to debts in Greece and other heavily indebted euro-zone sovereigns.

“At the very least these applications will undergo a much more stringent vetting procedure,” said Chris Rupkey, senior financial economist at Bank of Tokyo-Mitsubishi UFJ Ltd. in New York. “The lesson of history after these sudden financial bankruptcies is that regulators come down harder than ever. They don’t want to see another MF Global again on their watch.”

The Fed may need to increase the standards of its scenario analysis when determining the balance-sheet strength of the primary dealers, said Adrian K. Miller, senior fixed income strategist at Miller Tabak Roberts Securities LLC in New York.

A spokesman from the New York Fed declined to comment Monday afternoon.

The Fed had already tightened conditions for selecting dealers. The latest revision came in January 2010, which included an increase in capital requirements from $50 million to $150 million.

On that basis, some believe the Fed can’t be faulted for having one of its dealers go bankrupt.

“No one can hold the Fed responsible for the risk of the firm,” said Michael Franzese, head of Treasury trading at Wunderlich Securities in New York. “You try and put adequate procedures in place so it doesn’t happen but you have to trust people to do the right thing.”

Firms that have been seeking to join the primary dealer list include Toronto-Dominion Bank, the second-largest bank in Canada, and CRT Capital, a Connecticut-based broker dealer.

Membership has proven profitable at time when Treasury volumes have increased and demand for Treasury bonds has been fueled by the more conservative strategies of investors spooked by the euro-zone crisis and by the Fed’s support for the market.

After naming MF Global and Societe Generale to the list in February, earlier this month the Fed added BMO Capital Markets Corp. and Bank of Nova Scotia, both of Canada, boosting the dealer number to 22.

Now, with MF Global’s exit leaving a space open, other prospective dealers could campaign for entry, knowing that the central bank needs an enlarged pool to facilitate the giant bond transactions it undertakes to channel monetary stimulus into the economy, according to market analysts.

The Fed at the start of this month launched “Operation Twist” — a $400 billion operation to run until mid-2012 through which it will sell short-dated Treasurys and buy those with maturities of between six years and 30 years. Those transactions will be carried out via primary dealers.

Primary dealers will also be needed once the Fed starts to withdraw the cash it pumped into the banking system over the past few years, a strategy that has swollen the central bank’s balance sheet to beyond $2 trillion from less than $900 billion shortly before the 2008 financial crisis.

The Treasury Department has an interest in sustaining a large membership list for primary dealers, too. It needs these institutions to underwrite government bond sales as it continues to announce giant auctions to fund the fiscal deficit. If the auctions don’t go smoothly, the Treasury’s borrowing costs will rise.


  1. David,
    It should also raise awareness and concern.
    1. Who else is doing something similar? This is something that should have been expected with ZIRP. Reach through a combination of leverage and credit for more return.
    2. It also highlights potential links through the system via the collateralized repo market.
    3. It should cause the Fed, or at least some members, to question the efficacy of QE policies that cause these types of reactions, i.e., taking too much risk.

    This is the law of unintended consequences at work.


  2. Anon:

    1. Do you mean among the primary dealers?
    2. Could you elaborate on this point?
    3. Maybe. Not sure how much is "too much" risk though. They made a bet. They lost. That happens even with bets that have "just the right amount" of risk.

  3. David

    I'd don't quite see the big deal. A broker dealer failed.

    Big deal.

  4. Let me explain why I think the phenomenon is potentially interesting.

    I am studying a class of models which suggest that full and immediate disclosure of asset information on the balance sheet of an intermediary may not always be a good idea. Some nondisclosure may be optimal (with good behavior enforced by the desire to maintain the company's franchise value).

    Now enter the regulators/bond raters who force disclosure and downgrade to junk. Is this really in the social interest? Maybe yes, maybe no. I'm not sure. But I think it is worthwhile to think about. It is potentially a very big deal.

  5. How do you define "optimal", here? Seems it might differ for the shareholders / counterparties / employees etc.

  6. Anon, You are right to point out that interests may diverge here. But I meant optimal in the sense of Pareto, with a social welfare function consisting of an equally weighted sum of individual utilities.

  7. As I was reading this post I recalled the model you are talking about David (you taught it in a class). Essentially if people were unaware of the current value of MF's assets (if MF were not forced to disclose the information or perhaps just less frequently) investors may not have asked for their money back/more collateral to rollover debt/higher yields. This would have given more time for MF's bet to play out. Of course things could have gone even worse, but I can imagine a situation in which a little less information could have helped.
    Always like your take on things and the way in which you make your arguments.

    As for your point, Anon; it seems to me that no one did particularly well in this case (and if it's a pareto improvement, they are all better off -- weakly at least)

  8. Ben, Yep, you've had the benefit of listening to me describe the argument in some detail. I'm not entirely sure that it applies in this case, but it's always interesting to speculate "what if?"