Unwinding Drexel’s Swap Dealer After Bankruptcy
Wendy de Monchaux headed Drexel Burnham’s interest rate swap dealer when the parent firm filed for bankruptcy in 1990. The bankruptcy terminated Drexel’s swaps. Although transaction documentation specified the process, there were serious obstacles to unwinding hundreds of swaps with over 60 counterparties.
When Wendy came in to work at Drexel Burnham on Tuesday 13 February 1990, there was a rumor going around that her firm was about to file for bankruptcy. At 10:45 am the tape reported that Drexel had defaulted on a $100 million loan and at 11 am Drexel’s CEO, Fred Joseph, announced over the firm-wide intercom that Drexel appeared to be insolvent. Drexel officially filed for bankruptcy that evening at 11 pm.
Tuesday was the culmination of a three-year-long decline in Drexel’s fortunes. In 1986, Drexel Burnham made $545 million, making it history’s most profitable security firm. But unbeknownst to Drexel, that same year the US Justice Department began investigating its high yield bond business, headed by Michael Milken. Over 1987-88, aspects of the investigation gradually leaked out until in December 1988 Drexel pled guilty to six counts of mail and securities fraud and agreed to a $650 million fine.
But things got worse for Drexel in 1989. It bought back poorly performing high yield bonds it had underwritten. It got stuck holding a bridge loan to Paramount Petroleum when it couldn’t sell bonds to replace the loan. It also got stuck with West Point-Pepperell bonds it couldn’t sell. It was forced to guarantee Integrated Resources’ debt. (“Disintegrating Resources,” quipped the wags.) It suffered equity arbitrage losses in the failed buyout of United Airlines.
At the beginning of January 1990, Drexel couldn’t roll its commercial paper debt, which the rating agencies had downgraded. Later that month, Drexel reported a $86 million fourth quarter loss and a $40 million loss for all of 1989 and Citibank cut off its credit line. Drexel began upstreaming cash to its holding company from its broker-dealer to pay off holding company debt. In February, the Securities and Exchange Commission made Drexel stop the transfers. The day before Drexel filed, a hail-Mary meeting with bankers came to nothing.
DBL Products
Drexel’s bankruptcy not only put its employees out of work, it destroyed the value of compensation they had received for previous years’ work. A large portion of the five previous annual bonuses, denominated in shares of Drexel stock and held by the firm as deferred compensation, was now worthless. But unlike almost all Drexel employees, Wendy did not pack up her personal possessions and go home.
Wendy headed Drexel Burnham Lambert Products Corp. (DBL Products). She was one of few women in so senior a role at a security firm in 1990. Products specialized in interest rate swaps, where one party agrees to pay a fixed interest rate to another party agreeing to pay a floating interest rate, usually three-month LIBOR. These swaps might run for five or ten years. It wasn’t like buying a bond or stock, where after the transfer of security and cash the seller and buyer have no exposure to one another. Products’ swap customers were counting on Products to fulfill its side of the swap agreement for years.
Products was a separate legal entity from Drexel’s broker dealer, which held most of Drexel’s assets and did most of Drexel’s business. Products was not in bankruptcy; Drexel’s holding company had filed for bankruptcy. As a separate legal entity, it was technically still in business. Which is not to say that Products was unaffected by its parent’s bankruptcy. The agreements it had with its swap counterparties stipulated that holding company’s bankruptcy triggered the swaps’ termination.
This story goes into swap termination in detail, but here’s a quick first approximation of the process. After swaps are entered into and interest rates change, a transaction becomes one party’s asset and the other party’s liability. The market value of a swap is, roughly, the present value of expected net payments. If the expected future path of floating interest rates falls, the right to receive the swap’s fixed interest rate becomes valuable. If the expected future path of floating interest rates rises, the right to receive the swap’s floating rate becomes valuable. To terminate a swap, one party pays the other the market value of the swap. Wendy held the view that Products’ employees had an obligation to their customers to stay at their desks and try to accomplish swap terminations in as orderly and efficient a manner as they could.
It’s important to remember that back in 1990, interest rate swaps were un-intermediated bilateral agreements; there were no swap exchanges or swap clearing houses as there are now. Products employees’ task was to unwind hundreds of swaps with over 60 counterparties. No one had ever tried to do this. They took on the mission, but they faced daunting obstacles before they could even begin talking to their counterparties about swap terminations.
Hedging the Swap Book
First off, it was lost upon DBL Products’ bank that Products was its own legal entity, separate from Drexel’s holding company. In the immediate aftermath of holding company’s bankruptcy, Products’ bank, owed money by Drexel’s holding company and broker-dealer, impounded Products’ cash account and incoming payments to the account. Without the restoration of its account balance and banking services, it would be impossible for Products to terminate swaps. The other immediate effect of holding company’s bankruptcy was the liquidation of Products’ euro dollar future positions and financed Treasury security positions. Understandably, given the situation, The Chicago Mercantile Exchange and Treasury security counterparties did not want to trade with Products.
Losing the ability to transact futures and finance Treasurys meant that Products had no ability to use those instruments to hedge its swap book against interest rate moves and market value losses. If interest rates moved against Products’ unhedged book, it could become as insolvent as its holding company. Products’ risk managers began running the swap book every night through its risk management system to determine Products’ net exposure to interest rate changes. Products’ risk system expressed exposure to interest rates as a three-year series of three-month futures and then as longer-term Treasurys.
Under normal operating conditions, when Products could finance Treasurys and transact futures, Products’ risk managers would put on the opposite positions in futures and Treasurys. This would stabilize the swap book’s net asset value against interest rate changes. For example, if the risk system said the book was equivalent to being long futures and Treasurys, Products’ risk managers would short futures and Treasurys to get the book back to neutral. In actual practice, Products’ risk position was likely to be much more complicated than that; requiring long and short positions in different instruments and maturities.
So, the first task Products faced, before it could even think about terminating hundreds of swaps, was to make the swap book less sensitive to interest rate movements. The only method it had to do so was to terminate select swaps. But risk systems aren’t built to suggest hedges by terminating swaps. That’s not something any swap dealer would normally do to hedge their book. Products’ risk managers spent till 10:30 each night using the risk system’s futures and Treasurys results, intuition, and trial and error to figure out which swaps, if terminated, would best flatten Products’ exposure to interest rates.
Because swap dealers always stand ready to enter into trades, it was logical for Products to approach professional counterparties to terminate swaps quickly to try and hedge its book. Over three days, Products managed to take off enough of the right swaps with dealers to lower its book’s sensitivity to interest rate changes. Products wasn’t very well hedged, but it was a lot less interest-rate sensitive than after the sudden loss of its futures and Treasury positions. Meanwhile, using either legal reasoning, moral suasion, or threats, Wendy convinced Products’ bank to respect its rights, give it back its money, and allow it to make and receive payments from its cash account. Somehow, Products got through the first few days. Now its traders and salesmen could begin unwinding the swap book.
DBL Products’ Swap Book
DBL Products had three sets of counterparties with differing collateral arrangements. With other swap dealers, Products had two-way collateralization agreements. This meant that if the mark-to-market of swaps Products had on with a dealer was in its favor, that is, if it was an asset to Products, its dealer counterparty posted collateral, almost always US Treasurys, to Products. And if the swap was a liability to Products, Products posted collateral to its dealer counterparty.
Two-way collateral made it easy to terminate swaps with other dealers because it meant that the termination had little effect on the dealer’s liquidity or financing. Products’ dealer counterparty would either be giving Products cash and taking back its collateral, or getting cash from Products and giving Products back its collateral. From a combined cash and Treasurys standpoint, they were even, or nearly so, depending on how close collateral matched the swap’s mark-to-market. Terminating dealer trades would continue to be the means of hedging interest rate risk when Products began terminating customer swaps.
Another set of Products’ counterparties was federal home loan banks. Their swaps were under one-way collateralization, where Products posted collateral to the banks, but the banks didn’t post collateral to Products. The third counterparty group was comprised mostly of thrifts, which was DBL Products’ bread and butter business. On their swaps, Products typically had one-way collateral coming to it.
Under Products’ swap documentation with all its counterparties, the bankruptcy of Drexel’s holding company terminated swaps. Most of the swap documentation called for the termination to be conducted under the “second method, market quotation” procedure. This meant that when Products terminated a swap, it polled five dealers for quotations, threw out the highest and the lowest prices, and averaged the remaining three.
It’s important to realize that per its swap agreements, the quotations Products averaged were on its counterparty’s side of the market. This is another reason why it was relatively easy to negotiate terminations with other swap dealers. In 1990, there was about a five-basis-point spread in the interest rate swap market. That means a dealer would offer to pay a fixed rate on a swap five basis points less than the fixed rate the dealer would offer to receive on a swap. Instead of terminating swaps at mid-market, Products’ swaps were terminated at the side of the market that maximized the value of the swap to the counterparty. When Products terminated a swap, the swap dealer earned the present-value of the 2.5-basis point difference between its side of the market and mid-market.
Terminating Hundreds of Swaps
After DBL Products got its interest rate risk somewhat under control, its salesmen began talking to customers. Most customers were surprised to hear from Products. Salesmen heard “Are you still there?” a lot. Customers assumed Products’ employees had abandoned ship. The salesmen had to assure them, “No, we’re here and we have the ability to trade. We’re still operating. We clearly had a termination event under our swap agreement with you, and we are ready to unwind the trade per our contract.”
Sometimes salesmen were met with suspicion. Some customers had a fear Products was taking advantage of the situation. This put salesmen back a bit, because they thought they were performing a public service by staying at their desks and trying to clean up the mess. Products’ employees had nothing to gain; they weren’t getting their base salaries and they weren’t going to get bonuses at the end of the year, either!
But when people thought it over, they realized that if somebody at Drexel was on the phone, willing to talk and negotiate and get a deal done, that was a good thing for them. When they understood the termination procedures, they realized it was skewed to their advantage. Products’ best allies were often the counterparty’s own lawyers who told the transaction people that Products was doing what it was supposed to be doing under the terms of the swap agreement.
But additionally, people realized that if they could get out of their swaps and distance themselves from Drexel, that was a good thing. Having a swap on with the subsidiary of a bankrupt entity was recognized as not being a good situation. If there was one factor that facilitated Products’ ability to terminate swaps as effectively as it did, it was that counterparties were motivated to sever their relationship with Drexel. To those that needed a push to realize this, salesmen invoked a mantra. “You can either talk to me now and we can get this sorted out, or you can talk to the bankruptcy court in 12-24 months.” That often drove the point home.
Every day Products would try to unwind as many trades as possible, subject to the constraints of keeping its remaining book as interest-rate neutral as possible and managing its liquidity. With respect to liquidity, Products couldn’t freely terminate swaps that required outgoing cash payments. It had to balance incoming and outgoing cash payments and any incoming collateral. Each day they had a game plan where salesmen would talk to customers about the specific swaps to be terminated and traders would contact dealers to terminate swaps with the opposite interest rate profile.
So, Products began unwinding swaps. If the customer wanted to keep the swap, Products would bring in a swap dealer the customer already had a relationship with and assign the swap to that dealer. Then, Products and the other dealer would settle up the value of the swap calculated at the other dealer’s side of the market. If the customer was posting collateral to Products, Products would transfer that collateral to the other dealer. If the customer didn’t want the swap anymore, Products and the customer would exchange the value of the swap, again at the customer’s side of the market, and return collateral to one another.
In the half-hour or hour between terminating a customer swap and terminating the offsetting dealer swap, Products took the risk of a change in prevailing swap rates, small fries compared to being naked the risk overnight. On a great day, Products would terminate 20-25 trades and other days it would only terminate three. It took six weeks, but by the time Product’s employees left Drexel at the end of March, the book was unwound except for swaps with the few customers who would not engage with them.
One funny story involved collateral. One home loan bank was a huge counterparty with well over 100 trades. Products posted one-way collateral to the home loan bank. After one set of unwinds, someone sold a $14 million Treasury security Products got back from the bank to raise liquidity.
The problem was it wasn’t Products’ collateral. Products would typically pass collateral securities it received in from thrifts on to home loan banks. So, when someone sold the Treasury security Products received from the home loan bank and traders realized it belonged to a thrift, there was a panic. But a Products’ trader called someone at the thrift and without revealing anything said, “Hey, we’re going to unwind this huge block of business and we got a good side of the market. Why don’t I just sell the Treasury security you posted to us and send you cash, that’ll make the whole process nice and easy.” And the person at the thrift agreed!
Recalcitrant Counterparties
Besides the second method market quotation termination procedure, DBL Products had a few counterparties, mainly highly rated ones, under first method termination procedure. This means that the party not causing the termination has no termination payment obligation to the party causing the termination. Products had caused the termination because its parent had gone into bankruptcy. So, if the swaps Products had on with a counterparty under first method termination summed to a mark-to-market in Products’ favor, that was just too bad. The counterparty could just walk away and enjoy a windfall.
The logic of the provision is that such a counterparty is innocent and shouldn’t have to do anything if Products has a problem. Coming up with the net value of outstanding swaps might be a burden for such a counterparty. But on the other hand, in Products’ hedged book, what Products receives from one counterparty it owes another counterparty. One counterparty’s windfall causes another counterparty’s loss. In fact, the Financial Institutions Reform, Recovery, and Enforcement Act (FIRREA), signed into law before Drexel’s bankruptcy, made such walkaway clauses unenforceable in the case of an insured depository institution. But Products wasn’t such an insured depository institution.
Many counterparties with this “walk away” termination procedure paid Products anyway, perhaps because of the example set by FIRREA. But at least one counterparty persisted in its rights: in November 1992, a court upheld Midland Bank’s right to walk away from its swap liability to Products.
As Much as Could Be Unwound
This was the first time the swap market tested termination procedures with a huge counterparty like a swap dealer. Everybody came together to facilitate execution. Swap dealers were pretty good and brokers helped as well. Of course, it was easy for dealers to get involved because they got their side of the market when they took assignments or terminated swaps. You get a lot of focus when you are giving people money. Products got counterparties’ focus from their desire to get away from Drexel and the unknown. Those two things, swap dealers’ greed and end users’ fear, were the two drivers of Products’ ability to unwind almost all its book. Products was eventually incorporated into holding company’s bankruptcy and its recalcitrant counterparties had to contend with that process.
Products’ employees stayed around terminating swaps with counterparties because Wendy recognized it was the right thing to do for customers. But it also gained goodwill, which the swap team benefited from they moved en masse to Banque Indosuez and started business there. They were profitable there, but after three years Indosuez management decided they didn’t need Wendy and let her go. When she landed at Bear Stearns, most of the team left Banque Indosuez and joined her at Bear, so a lot of the original Drexel team worked for Wendy at three different security firms.
Publisher’s Note: Wendy de Monchaux told me this story over 30 years ago. She passed away suddenly in 2017 after suffering a seizure, leaving behind her husband, David MacWilliams; three children; and the many friends and colleagues who held her in high regard. Mike Fedak, who co-headed marketing at DBL Products, provided his perspective of events. Errors in the story are the result of my poor memory.
Wendy became global head of derivatives at Bear Stearns and a director of Bear Stearns International. A Wall Street Journal obituary suggests the remarkable woman she was.
To comment on a story or offer a story of your own, email Doug.Lucas@Stories.Finance
Copyright © 2023 Douglas Lucas. All rights reserved. Used here by permission. Short excerpts may be republished if Stories.Finance is credited or linked.