The first recorded instance of the term “fire sale” is thought to come from an advertisement posted by Maraton Upton, a resident of Fitchburg, Massachusetts, who had suffered through a massive fire at his home in December 1856. Following the destruction, Upton no doubt needed money for recovery, so he took out an advertisement that read in part, “Extraordinary fire sale; customers are invited to call and examine goods which are still warm.”
These days fire-sale is a term used in many facets of American life. Regardless of the setting, however, the meaning is almost always the same: something is for sale, condition notwithstanding, for significantly less than it would cost new. For that matter, not only is it less than a discounted price, the term implies the items are being sold at a drastically reduced price.
Financial Market Fire-Sales
In the financial world, fire-sales often occur after an entity defaults and its counterparties are forced to sell-off its positions. The Fed is currently worried about the impact of fire-sales on the financial markets in what they call a “collective action problem” – if everyone sells at once, it drives prices down even further than if everyone coordinated their sales. It’s kind of like the “prisoner’s dilemma” for financial markets, where the first one to talk (sell) can get themselves a better deal. But if the prisoner’s (dealers) all cooperated, they’d get themselves a better deal together as a whole. The Fed is pushing the idea of an industry consortium to liquidate defaulted counterparties in the Repo market, specifically in the Tri-Party Repo market.
Long Term Capital Management – The Success of A Dealer Consortium
In one respect, the fate of LTCM can be viewed as a successful dealer consortium that resolved a default. In the case of LTCM, all* of their counterparties contributed funds for an orderly wind down. When LTCM collapsed in 1998, not only did the Street have massive amounts of exposure to them through Repo financing trades, but there was also correlation risk with the big dealers’ proprietary trading positions – they all had many of the same trades on. It was certainly in the Street’s best interest for LTCM to be wound down in an orderly manner, otherwise they’d all be liquidating LTCM at once. When the Fed hosted the bailout meeting, the interested parties agreed to contribute $3.625 billion to fund the rescue. After the liquidation was complete, the industry even made money on the bailout. There’s a good argument to be made for an industry consortium taking over the failing entity and winding down their positions in an orderly manner.
Amaranth – The Uncertainty of Bankruptcy Court
In 2005, a little known natural gas trader named Brian Hunter at a hedge fund named Amaranth Advisors made a big bet that natural gas prices would increase for the Winter 2005-06 natural gas season. After hurricane Katrina hit followed by hurricane Rita, it knocked out natural gas facilities and pipelines in the Gulf of Mexico. Hunter and Amaranth made a killing, they were up over $1 billion on the trade. Figuring the same thing would happen the next year, Hunter placed an even bigger bet on natural gas futures and OTC swaps in early 2006, controlling as much as 60%-70% of the winter delivery season. The 2006 hurricane season was a dud and Hunter’s positions were so large, Amaranth could not get out. They managed to hold on until the middle of September as natural gas prices were in a free fall, but they eventually succumbed to margin calls. Around September 15, Amaranth was down to net assets of $3.5 billion, after having lost over $5 billion so far that year. They struck a deal with Goldman Sachs to buy their entire energy portfolio, but there was one catch, JP Morgan, Amaranth’s futures clearer, had loaned Amaranth money against their futures positions.** JP Morgan had a claim to the assets, which meant funds could potentially be “clawed back” in bankruptcy court years later. Faced with the uncertainty of an unfavorable ruling in court, Goldman backed out of the deal. Soon after, Amaranth’s positions were sold to a group consisting of JP Morgan and Citadel for the sum of $2.5 billion.
The bottom line is that no one wants to be tired up in bankruptcy court and no one wants an unknown potential liability down the road. When there are large degrees of uncertainty, market participants pull out of a market or increase their bid/offer spreads to reflect the potential of additional costs.
BSAM – The Problem of Holding Positions Too Long
The CDO market in 2007 is the perfect example of contagion that can be caused by fire-sales and the downward spiral of prices. Leading up to 2007, the CDO market had already peaked and began to decline sometime in 2006. Some dealers were more aggressive than others in market down their customer’s CDO positions, but eventually the entire market was eventually forced into decline. As margin calls increased and forced sales pushed prices lower, one leveraged fund after another collapsed, forcing more sales and pushing CDO prices even lower, which then forced more leveraged players into default.
One of the domino’s that fell in June 2007 was Bear Stearns Asset Management (BSAM). Merrill Lynch was one of their Repo counterparties who were forced to take over the billions of BSAM’s CDOs that Merrill Lynch was financing. Merrill made a catastrophic mistake in the liquidation – when they couldn’t sell the positions at prices they deemed acceptable, they held onto the paper. Down the road, they were force to take even larger losses and some their CDO paper was held until it was worthless.
If there was an industry consortium in charge of an orderly BSAM wind down, there’s a good chance they would have done what Merrill did and held onto the positions, especially if traders at Merrill were recommending that strategy. Getting back to the downward spiral of prices, had some industry consortium liquidated failing entities, does anyone honestly think those leverage funds would still be alive today had they not collapsed in 2007? In this case an industry liquidation consortium would have just slowed down the decline of the market and spread losses around the entire industry, away from institutions, like Merrill, which had made bad business decisions.
Lehman – The Success of Central Clearing Party (CCP) Liquidations
Though Lehman is blamed for causing financial havoc across the globe, unwinding Lehman’s Repo book was not a large problem. Lehman was a member of LCH.Clearnet in Europe and Fixed Income Clearing Corp (FICC) in the U.S. After Lehman declared bankruptcy, both CCPs liquidating Lehman’s Repo positions and neither CCP reported a loss or called on their default fund. Repo liquidations worked very well under the extreme market conditions of September and October of 2008. However, other parts of the Lehman liquidation did not go so well, in fact, Lehman is still being wound down almost 6 years later.
MF Global – The Problem With A CCP Liquidation
Before MF Global declared bankruptcy, the “cat was out of the bag,” so to speak, and market participants knew MF Global would be liquidating their Sovereign bond portfolio among their other positions. After the bankruptcy was declared, LCH.Clearnet assumed control of the portfolio and arranged an auction. For one of the buyers, George Soros, the purchase was extremely lucrative. In total he purchased approximately $2 billion worth of the $4.8 billion RTM (repo-to-maturity) securities that were offered for sale. For the Italian bonds, he paid approximately 89 cents on the dollar, compared to the market price of 94 cents and within a month those bonds were back up to a price of 96, making Soros a $140 million profit in the space of a very short period of time.
There is, of course, no way of knowing the full extent of all the losses due to the “fire sale” of MF Global and certainly under better circumstances a liquidation could have been handled better. Looking at the numbers, we can surmise the approximate damage done to MF Global’s bottom line. For starters, the firm took a $7.3 million loss on the sale of $600 million worth of RTM positions before the bankruptcy. The sale of commercial paper to Goldman Sachs for $1.33 billion had resulted in a loss of $15 million. Soros had paid $2 billion for a portion of the RTM portfolio, costing MF Global another $120 million, with the remainder of the RTM positions selling at a 10% discount, which adds another $280 million to the loss ledger. Moving on, we come to the $418 million worth of corporate securities the firm held, which were sold for $384 million, a loss of about $34 million. An additional $600 million worth of relatively illiquid securities were on the books, which we can conservatively guess were sold at a 10% discount, meaning they lost the firm another $60 million.
Assuming a negligible loss or even a slight profit realized on the sale of the remaining assets in the firm’s holdings – which included several billion dollars in federal agencys and U.S. Treasurys – the total cost of unwinding MF Global’s trading positions in a one week fire-sale topped out at over $500 million.
Problems with a Dealer Consortium
The dealer consortium worked well during the LTCM crisis, but it wouldn’t have necessarily work well under other markets. If you know the market will remain stable and eventually return to normal conditions, the dealer liquidation consortium is a good way to go. However, if the market will continue to decline, a consortium is a bad idea.
You can say the right to a quick liquidation is what makes the Repo market possible in the first place. It’s pretty clear that if LTCM’s Repo agreements had specified that: “In the case of LTCM’s failure, you will join a consortium of banks and invest $300 million of your own capital while the fund is wound down,” then LTCM never would have been in business in the first place. The right to liquidate Repo transactions quickly and easily in the event of a default makes Repo financing possible and is responsible for the size and efficiency of the Repo market today.
Another problem with the dealer consortium is the potential conflicts of interest it creates. If the chickens are disappearing from the hen-house, it probably doesn’t make sense to let the foxes take turns guarding them at night. Just before LTCM went down and the fund was frantically looking for a buyer, Jon Corzine’s traders at Goldman were seen downloading LTCM’s positions onto a laptop computer. Many firms involved in a liquidation could also benefit from the liquidation.
There’s another risk too: an orderly unwind requires someone to make market trading decisions. Is it better to hold the securities for months or sell them now? Collective industry solutions involve taking market risk and a view on whether the market will “recover.” That worked in 1998 for LTCM, but it would not have worked in 2007, just ask Merrill. As the months went on, Merrill watched the CDO market continue to deteriorate and eventually took billions of dollars in losses.
Then there is the issue of the bankruptcy court. Any “trading” decisions that are made to hold securities means there’s potentially a liability in bankruptcy court. If market prices deteriorate and someone made the decision to ride the losses, then the Trustee is coming after those losses. Liquidation trading decisions open the door to potentially millions or billions of unknown liabilities down the road.
Hosting A Fire-Sale
I had the opportunity (not sure if that’s the right word) to liquidate a large Repo customer once. It was a well known name, and of course, it was right in the middle of a financial crisis. Obviously, there’s no coincidence these things happen at the worst times. This customer was leveraged with well over a billion dollars in corporate bonds, structured corporate bonds, and some CDOs, plus small amounts of U.S. Treasurys and federal agencies.
One of the hardest parts of a fire-sale liquidation is that you’re unsure of the outcome. As soon as you legally take over the customer’s positions, you’re in a no win and possible lose position. If you sell the positions and there’s excess cash left over, you must return the funds to the estate. If you’re sitting on a a loss after the liquidation, you then stand in line at the bankruptcy, not expecting a settlement for years. When the bankruptcy does get resolved, you’ll probably receive a fraction of what you’re owed.
So the key to the liquidation rests on not losing money when the customer positions are sold, and that rests on the mark-to-market prices and the margin you’re holding. If the prices and margin are sufficient, there’s a good chance of getting out of the liquidation unscathed. If the mark-to-market prices are inaccurate and the margin doesn’t cover the movement in market prices, you’ll be staring at a loss. What’s worse, you don’t know the size of that loss until the liquidation is complete.
Here are my recommendations for handling a liquidation:
- Devise A Plan – Determine how much to sell each day, the maximum loss you’ll accept on a security, who you’re going to speak with, and who you trust.
- Counterparties – Determine who is the best to bid for the different types of paper. For example, some banks are better for CDO prices and the individual banks should be the best bid on their own paper.
- Keep A Low Profile – The key is to not to let the Street know the customer’s positions and their size. If the Street gets access to too much information, they’ll front-run you.
- Stick To The Plan – When there’s a loss or prices are lower than you expect, sell them and don’t try to trade out of a bad market. Holding can work sometimes, but as a rule of thumb, it’s worst strategy. As we’ve seen, holding positions will also open up the liquidator to questions from the Trustee years later.
* Except one, Bear Stearns, but that issue is debatable considering Bear was LTCM’s clearing firm and had potentially more liability than the other firms.
** Called “margin finance” in the futures industry.