Print this
Confessions of a Structured Note Salesman

An anonymous insider tells a story of ignorance, greed and creative frenzy during the spectacular rise and fall of the structured note market.

''In 1991, I was working as a quant, selling CMOs at a commercial bank. I was feeling miserable because I thought my career was in a slump. Once a quant, always a quant, or so I thought.

I was about to look for another job when I got called into my boss's office. He told me they were starting this new group elsewhere in the firm and that my name had been mentioned. It was a new opportunity in the bond market that married derivatives and bonds. I thought derivatives were cool, so I joined.

In mid-1991, the structured note market was a very quiet and privately negotiated market. We were doing very simple structures at first, floating rate notes, or notes based on Libor or Prime that paid a higher coupon than normal because the return was capped at a certain level. Occasionally, however, we'd get an inquiry at the swap desk from someone like Lehman or Merrill, asking us to do these weird funky structures tied to callable bonds. We knew something was going on but we couldn't quite figure it out.

The party begins

Then around the summer of 1992, public deals began showing up on Telerate's Corporate Watch page. The bigger shops like Lehman and Merrill, Solomon and Goldman were bringing in $50 million Sallie Mae deals. That was when I saw my first inverse floaters* and leveraged-capped floaters* - which seemed pretty curious at the time.

These notes were issued by the agencies and some corporates off of their medium-term note programs. They would get a reduced funding - sometimes as low as Libor minus 60 or more basis points. By the fall, the Federal agencies like Fannie Mae, Freddie Mac and the Federal Home Loan Banks started entering this market in force, and spreads narrowed to 45 basis points.

Although the notes would pay investors based on some arcane formulae, the issuers were never exposed to this complexity. They didn't even have to understand them. Swap dealers attached a swap to every note so the issuers were hedged out and simply paid their target fixed or floating rate cost of funds.

The investors buying this debt were either short duration fixed income funds, basic core bond funds at mutual fund companies, pension money managers or bank trust departments. The issuers on the other side were finance companies like GECC, GMAC, Heller and Household Finance. They are all ready takers of funds as they were always raising money to finance their loan portfolios.

It was an unusual opportunity, in retrospect. The yield curve was steep and getting steeper as the market anticipated higher future rates. People who thought that rates would trend down, however, could take a position against the market, betting against the forward rates and earn an enhanced coupon. Regular floaters were generally paying 3 1/4 percent at the time. But if you bought an inverse floater, you might get a coupon of 7 to 8 or 9 percent, depending on the gearing of it.

It all seemed like a safe bet at the time. The Fed had been very accommodating. There was talk about the unwritten commercial bank bailout that the Fed engineered by keeping short-term rates at 3 percent. Banks could borrow overnight at Fed Funds and then invest in ten-year Treasuries at 6 percent. They would do this trade in huge size as the positive spread allowed many banks to rebuild their capital.

Our shop was late getting into the structured note business and slow in catching up. We started selling very simple structures at first. But other firms in the market started creating lots of new structures - like Step Up Recovery Floaters,* leveraged CMT floaters and their variations. There were also notes based on the difference between two indices - say the CMT rate minus Libor or CMT minus Prime. The expectation was that the spread between the two interest rates would widen and an investor would get a dramatically enhanced coupon.

Merrill and Lehman were the dominant players in 1992, but it crowded quickly. Goldman, Solomon, Bankers Trust, JP Morgan - any firm that had distribution and derivatives capability - was active. In my shop, one of the largest US commercial banks, we were doing a tenth of the volume of the others.

The three types

In my mind, I grouped investors into three basic categories. First there were the simpletons. They'd say "Wow, I can get 3 percent over there, but I can get 6 percent over here. I don't really understand it, but I want it." They didn't even ask how they were getting that coupon. They just wanted to buy something that would go up when rates went down.

Second were the selfish investors who understood what they were doing but had their own agenda. Maybe their agenda was yield enhancement, maybe it was their bonus. They figure they'd buy a structured note and hold it for a short time while earning the above market rate; it would help the fund's performance. If you're a mutual fund manager, you get a salary and a bonus based on your performance and on how big your asset base grows. If you could buy a bond that could give you an above market coupon, you could report in your newspaper ad that your current yield was higher than your competitor's. New money would flow in and then you'd collect a big bonus. These guys were smart. They were in and out of these trades in six months or less.

The third class was the smart, sophisticated players. They really knew what they were buying and wanted to buy it because they had an economic rationale for taking a position in the market or because they could not use OTC derivatives and had no other way to put on the trade. I didn't talk to many people like that. I don't know if my experiences are representative, but there seemed to be a lot of accounts that I would put somewhere between the simpletons and the selfish.

A lot of investors were getting snowed. That's not to imply that whoever sold to them was being fraudulent. I can't speak for how everyone marketed the product. It's clear, however, that some product was misrepresented and people didn't know what they were buying. What was a short-term cash fund of a municipality doing buying a leveraged inverse floater? They just didn't understand the whole concept of projecting forward rates and using the forward curve to value these things.

Betting against the forward rates

We would talk a lot to investors about what the market might be implying, which can be expressed in terms of forward rates. Forward rates are a construct. You can trade them, but forward rates are not a good predictor of future rates.

Most investors knew that they were using structured notes to take a punt on the market, but it was a punt they felt comfortable with. If they were buying a floating rate note capped at 9 percent, they'd say, "I don't think rates are ever going to go to 9 percent so I can buy the cap floater and get an extra 20 basis points of coupon." Or you'd say to them: "Given where Libor is today (3 percent), Libor would have to rise by so many basis points per year in the next three years for you to lose money over buying a regular, vanilla floater."

Of course this is the wrong way to value these notes. Instead of making a decision on where you think rates are going to go, you should value structured notes off of the forward rates - just as you would swaps. If any investors did this, they would have realized they weren't getting good value out of the trade.

The problem was particularly acute when you purchased a multi-index floater* - such as a note based on CMT minus Libor or Prime minus Libor. They aren't what they look like. These trades appreciate in value as the yield curve gets steeper, but they lose value more quickly as it flattens. Some people thought these things would always trade like regular floaters, which are reset every quarter. But they really behaved a lot more like an intermediate-term bond.

Principal-linked deals

The early structured notes were coupon-based. If you put in $10 million, you'd get $10 million at maturity, but the risk was in how much coupon you would get. In the fall of 1992, however, I started to become aware of principal indexed structures, where your coupon might be fixed but your principal varied according to some formula. These trades, of course, were a lot riskier than anything we'd ever seen - particularly when they were leveraged.

One of our salespeople got a printout of a client's investment portfolio. I looked at this printout and saw all these one-and-a-half year and two-year securities, medium-term notes issued six months ago by companies like GECC, CIT, Honda Motor Credit and Household Finance. But they all had these coupons of 8 percent or 9 percent! We all stood around trying to figure out, "How did they do that? How do you get an 8 percent coupon out of a two-year deal?"

Then, a little later, a client sent us a term sheet on one deal where the principal at maturity was indexed to where two-year swap rates might be. After sitting down with it for a little while, you draw boxes and figure out how it's done. It wasn't too difficult. Then we were smacking ourselves on the head. Oh! Of course!

I think that particular note was from Bankers Trust. BT was always an innovator and everyone knew that. They're still an innovator. At that time they were coming up with new plays on the market that competitors wouldn't even think about for a year. We couldn't figure out how to structure some of them, and even when we did reverse engineer them, we couldn't trade the risk at my firm because we didn't have the models.

Climbing the curve

Every swap house has certain types of risk they can trade. To create a particular type of product, you need to have certain manufacturing capabilities. You may understand that a Ferrari is a beautiful car, but if all you can do is make motorcycles, you won't know how to build it. Our shop built bicycles and figured out how to put a motor on them. Then we'd see one of these Ferraris go by and we'd be absolutely blown out of our shoes.

Eventually, however, we'd figure it out. Once we saw what they were doing, we'd start asking, "How can we price that?" And we'd conclude: "We can't." So our traders would go off with a quant guy and rip apart some computer code and figure out a way to price it. There was huge money to be made trading these different kinds of products.

Things really started to mushroom in late 1992. That's when the retail and smaller regional dealers started to get involved and the distribution became much broader. The bulge firms were manufacturing more product than they could sell. So you had all these other dealers who saw these products as great revenue opportunities. Firms like Dean Witter, Paine Webber, Smith Barney, First Tennessee, Gruntal and some small outfits in Texas like Government Securities Corp. and MMAR Group.

Merrill would get an anchor order from an investor on a particular structure for maybe $25-50 million and they would get the regional dealers to buy $5 million and $10 million each. Before you knew it, you had a $100 million deal printed. These regional dealers would then break down the $5 million they bought into half-million dollar lots and sell them to their clients. A few months later they realized that they didn't have to buy from Merrill et al. and went to the swap houses and agencies directly to manufacture their own deals.

The international market opens

In the beginning of 1993, we started to see international plays. International bond funds began to use structured notes as a way to make new kinds of plays in Europe. The thinking at the time was that the European markets were six months to a year behind the U.S. in terms of the economic cycle. The yield curves were also very steep throughout Europe, so you could make the same kinds of bets against forward French Franc-Libors or Deutsche mark-Libors and have the proceeds paid to you in dollars, with the currency fluctuations hedged away.

We also started to see exotic option structures. There were one-way floaters* that would pay enhanced coupons as interest rates rose but guaranteed investors a certain minimum return if they fell. Then there were notes with embedded binary options called range floaters,* where you'd get a high coupon if interest rates remained in a certain range, but you'd get nothing if they didn't.

Everybody was selling product to clients at whatever sector of market they were in. It was all very exciting because we were all making money. Everyone had a reason to be happy. The investors were excited because they were getting something they wanted - a great coupon. The issuers were getting cheap money. The regional dealers were making money, the swap houses were making money and the large underwriting firms were boosting their league table standings.

A well-oiled machine

After a while, it became a well-oiled process at our shop and at the issuers,' too. The Federal Home Loan Bank had an assembly line down in Reston, Virginia. On some days in 1993, Federal Home Loan Banks were issuing a dozen deals totaling several hundred million dollars. One day, I recall, they did over $2 billion. If you wanted to do a deal in the height of the structured note market, you'd have to queue up every morning to get your deal executed. You'd fax in the term sheet and then ask, "How busy are you?" They'd say, "Well, there are three deals ahead of you."

So you'd wait your turn to get on the conference call on the recorded line with the regional dealer and the swap house - the purpose of which was to confirm details of the trade. The FHLB would begin, "We have so and so on the phone, and this trade is a $20 million multi-coupon step-up note"...and then they would literally read through pages and pages of the bond and swap term sheets: "The issue date is this, the maturity is this, the coupon is semi-annual..." On a ten-year deal, you could have 20 coupons. It went on and on. "The coupon from September 1993 to March 1994 is 6.5 percent, the coupon from March 1994 to September 1994 is 6.75 percent..."

I remember doing one long-term multi-coupon step-up* deal when the market was choppy and moving against the trade. Halfway through the reading of the coupons, one of the lines on the desk rang. I was totally confused. It was the dealer who I'm already on the conference call with! He gets on the line and says, "Where am I now? How's my fee?" He knew the market was moving away from him, so he called me on another line to ask me how much money was still in the trade. While we were reading the numbers on the conference call, the bond market had ticked up a couple of thirty-seconds. That cost him about 10 percent of his fees. There wasn't a whole lot of money at risk, but if you're making $175,000 on a deal and you're watching $20,000 slip away, that's a reasonable amount of money. I told him how much his fee had moved and he had to eat the difference. The deal got done nonetheless.

At some point, however, the innovation process became self-perpetuating. People were innovating for the sake of innovating. I would liken it to what I saw in the mortgage market when I was involved with CMOs. Why did anyone ever buy a "Jump-Z" bond anyway? You can always say, "We're going to try to find a way to make more money or to get an enhanced yield." But at some point you have to stop and say, "This has gone too far." I think that's a purely subjective call. Everyone draws their line in the sand differently.

I've seen things in the market where I scratch my head and can't imagine why people did it. For example, when P&G lost all that money, I couldn't fathom what anyone at the company was thinking when they looked at the formula of that swap and said "Yes, that's exactly what I want to put on."

The party flags

Toward the end of 1993, the party was starting to end. The U.S. long bond hit a low sometime in October, but the yield curve started flattening and investors started to talk about when the Fed would raise rates.

The inverse floaters were the first to lose value as the curve flattened. A lot of money funds, fearing that they would "break the buck," started to sell their paper. There was no developed secondary market then. A lot of times, an investor would go back to the dealer who sold him the note, but the dealer wouldn't give him a bid, or would give him a terrible bid. You could go back to the issuer and try to get them to unwind the deal. If it was a Home Loan Bank, you could try to get it recut. But recuts were a paperwork hassle because they'd have to tear up a bond offering and write a new one - and they wanted to get 10-20 basis points for their trouble.

I found that we couldn't sell what we used to. We were still making the same arguments as before, but they were nowhere near as compelling because people weren't as confident that rates would continue going down. After Thanksgiving of 1993, the curve started flattening in Europe, too.

We had to do a daily mark to market for our mutual fund accounts. Around December their profits contracted. I recommended that they take their profits and get out - because economic activity was picking up quicker than expected. But I didn't ever stop and think the whole thing was going to blow up.

Out go the lights

Then the Fed finally raised rates 25 basis points in February. The sirens and fire alarms went off and I thought it was officially "The End Of The Bull Market." The lights had been turned on, everyone stopped necking, and the party was over.

The ensuing selloff had as much to do with the popping of the leverage bubble as anything else. My accounts weren't the type that were running for the doors. But valuations changed a lot over the course of a few days. Long bonds sold off a tremendous amount. I think what drove all this was essentially a readjustment of peoples' expectations of lower inflation. That was coupled with a whole lot of selling to get out of leveraged yield curve positions in the market.

With all of this selling it didn't take us too long to figure out that we were going to be actively buying paper. There were lots of structured notes looking for bids or unwinding. We would take apart the trades and repackage them into plain floaters that we sold back to banks or even to the agencies that issued them. This became the market for destructured asset swaps* - not a bad deal, eh? An agency issues at Libor minus 40 and then buys back its own paper or another agency's at Libor plus 10. That gives you 50 basis points a year for AAA risk!

The majority of '94 and even now continues to be undoing what was done. The exotic market has pretty much ground to a halt. You don't see anyone taking these outrageous directional bets in the market like they used to. Everything is back to simple again.

Anyone who is in the category of investors who have been burnt publicly in the press is pretty much out of the market. Portfolio managers can lose their jobs if they buy a structured note now.

There's been a lot of erroneous comment about how much money was lost in the market. Anybody who was long in 1994 lost money. If you owned a Treasury or structured note you lost money. But what people say is, "I lost money on a structured note and it was because of derivatives." But that isn't necessarily true. In fact, you may find that more money was lost in vanilla products than in structured notes. Although there was often some leverage factor in these notes, the effective duration tended to be shorter than five years. Remember, too, that a lot of people who lost money in '94 made an enormous amount in '92 and '93.

This year we're starting to see some accounts coming back. But they're the more sophisticated players. I observe a substantial increase in sophistication. They want to see scenario analysis now, and break-even analysis. I think that's very good.

More funds will come back into the market this year or next. Some have modified their prospectuses to allow the direct use of derivatives. But for most investors there's still a long cooling-off period that we'll have to live through. If you look at all the swap houses, you'll find that revenues are off everywhere. It's gonna be a tough couple of years - no doubt about it.