Q: The development of derivatives is largely driven by investor needs. What are some of those needs? A: A large number of investors are starting to realize that they face a huge reinvestment risk in the middle of the decade. If you remember, municipal issuance peaked in ~85 and ~86. Municipal bonds tend to be callable 10 years after they're issued, so we're just starting to see the beginning of a wave of redemptions. And that creates serious problems, particularly for some property-casualty insurance companies and for mutual funds. Q: Let's start with the insurance companies. How are they affected? A: Insurance companies reporting book income to their shareholders right now still include coupon income on higher-coupon bonds that they bought back in '84, '85, '86. They're reporting that income, and it goes into their per share earnings, and, multiplied by the multiple of the stock price, affects their stock price. Those earnings will disappear starting in ~94, ~95, ~96, ~97, and so on. A lot of insurance companies are beginning to anticipate some fairly substantial drop-offs in per share earnings in those years.
Since the equity analysts on Wall Street tend not to look very closely at an insurance company's bond portfolio in analyzing their earnings -- they tend to look at the underwriting results -- these declines in earnings will be a surprise to the Street and could have an adverse effect on stock prices. When you look at some of these companies, the numbers are pretty large. Q: How might an insurance company deal with this problem without using derivatives? A: If you have a lot of bonds maturing or being redeemed in ~94, ~95, ~96, you could sell them and buy longer-term bonds. That is a legitimate being to do. A lot of companies are considering it and some of them are doing it. But it creates two problems.
One is that you're realizing a gain. You bought the bond at par and you are selling it at 120 or 125. So you will have to pay capital gains tax. For a corporation, that means 34% capital gains tax on those realized gains that you wouldn't have to pay if you just kept the bonds.
Second, insurance companies are now starting to face more stringent standards regarding mark-to-market accounting. It used to be that they could sell bonds and not have to mark their portfolio to market. Now the standards have changed. If they do a lot of selling, they run the risk in some cases of being required to mark their entire portfolio to market. Q: Even the bonds they're keeping? A: Even the bonds that they're keeping. Right now, that wouldn't necessarily be a problem. Everything they own is at a gain and they would realize a large accounting gain for some purposes. But they don't want to be caught in that requirement going forward in case the market sells off. So there is a cash solution but it's an expensive one and, from an accounting point of view, perhaps a risky one. Q: So we turn to the world of derivatives, I presume? A: One solution that a lot of companies have looked at and some of them have actually got involved with is municipal forwards. Issuers who could not do advance refundings but like the market now will issue bonds for forward delivery. Q: How would it compare with the cash trade? A: It doesn't have either the tax or the accounting disadvantages of the cash trade, of actually selling the bond. The insurance company would keep, or the mutual fund for that matter would keep, their higher-coupon bonds, continue to earn the higher income, not pay the capital gains tax, but would schedule a forward purchase at about the time their prerefunded bonds are being redeemed. So they would just get the proceeds of the redemption, and use those proceeds to pay for the forward bonds on the delivery date. There is no sale, so no realization of gains. No sale also means no question about possibly having to mark an entire portfolio to market.
But it's not much of a solution because there's not enough issuance [of municipal forward contracts]. If there is more, than that will become a bigger part of the solution. Q: What else could investors do with derivatives? A: Another derivative type of solution that people have looked at and some have used is the use of inverse floaters -- INFLOS as we call them, or Indexed INFLOS. If I know that I'm going to be getting a lot of money two years from now as proceeds from redemption, that's only a problem for me if interest rates stay low or go lower. If interest rates shoot up between now and then, then I don't have a problem. My earnings will remain high because I'll be able to reinvest at a higher rate. So in a sense, if I'm in this situation, I'm short the market, I'm hoping that interest rates go up. I've got a bet on that, in fact, implicitly that interest rates will go up. Q: It's like owning a lot of two-year bonds? A: Exactly. And so I may want to balance that risk by buying some bonds that will do very well if interest rates stay low or go lower. And that's just what an inverse floater will do. In a sense, what portfolio managers in this position want to do is extend portfolio duration, to lengthen their portfolio. Q: Give me a quick definition of duration. A: Duration is the weighted average timing of discounted cash flows. It's also the price sensitivity of a security or a portfolio to interest rate changes. It's related to but not identical to maturity.
Portfolio managers are trying to extend the average maturity of their portfolios and they can't do it through cash trades because of these cash and accounting constraints. One way that they can do it efficiently, getting a lot of duration per dollar invested, is to use some of their current cash flow to buy inverse floaters. Q: Any other alternatives? A: Another way of dealing with the problem is a security something like the one that was described in The Bond Buyer a few weeks ago, an embedded forward. It's a municipal bond with a coupon that will step up or float upwards if long-term bond yields decline and will go down if long-term bond yields go up. It's another type of security that an investor can buy which will be an efficient way of extending portfolio duration. Q: If investors decide to use inverse or embedded floaters as a hedge, what percentage of their portfolios should they make up? Should they go with 5% or more? A: I don't think a percentage of anything is the right number. I think what they probably want to look at more precisely is the target duration they want to have, or what target average interest rate sensitivity they want to have. And then look for the most efficient way to get to that target interest rate sensitivity.
In other words, if I have a portfolio that's 70% [prerefunded bonds], and 30% 15-year bonds, that's a fairly short portfolio from a lot of people's view, especially considering the balance sheet considerations that effect a lot of insurance companies. If that were the case, and I didn't want to sell anything, and everything was at a gain in my portfolio, then I could see using all of my marginal cash flow for a while to buy inverse floaters or step coupon bonds or forwards just to get more in balance.
The relevant characteristics of a portfolio are thinks like duration and exposure to call risk and so on. How you achieve that is not by having hard-and-fast rules. Q: Should people have a preference with regard to using a more or less volatile derivative? Should I use a five-times leverage inverse floater or buy one with less leverage? A: There are arguments on both sides. The more unusual a derivative is, the less liquid it will be. We've done a lot of trading of what you might call vanilla derivatives. We've bought and sold in the secondary market our own floater/inverse-floater product, and we have bought and sold in the secondary market both of our main competitors' inverse-floater products. We've developed a fairly liquid market for vanilla inverse floaters under normal circumstances. So that is an argument for using something that is one-to-one leverage.
The argument for using something with more leverage is that from just the point of view of achieving portfolio objectives quickly and efficiently, you can do it faster with more leverage. The other argument for it is that issuers charge, through lower yields, higher prices for their bonds on the initial offering. It is sort of a fixed amount, regardless of how many derivatives are piled on at the end of the transaction. So you wind up paying issuers less per unit of the portfolio characteristic that you're buying if you have more leverage.
Leverage cuts two ways. The amount of leverage that you can put onto one of these trades is limited by the fact that the coupon that you receive, after all the calculations are done, can't go below zero. If you have a very highly leveraged instrument and the trade in a sense goes against you, you use up your coupon very quickly. That is unfortunate but it also has a pricing implication. It means that you, the buyer of the inverse floater, in effect, are also buying a cap on the J.J. Kenny [short-term index] or the [Public Securities Association Swap] index. If you put a lot of leverage on that, that cap becomes very valuable. And that will affect the efficiency of the transaction. You may wind up with diminishing marginal duration per unit of leverage very quickly. Q: We talked about the insurance companies with a particular need. Are there other kinds of investors with different needs? A: Other types of portfolio managers have the opposite risk. They're at risk if interest rates shoot upwards. I put the leveraged closed-end funds in that category. The dividend on their common stock could be jeopardized if short-term interest rates spiked up. Q: Could you you give me an example? A: A leveraged closed-end fund, for example, might go out and issue $100 million worth of common stock and buy long-term municipal bonds with it. Then the fund might issue $50 million worth of preferred stock and buy more bonds. The preferred stock rate is reset through an auction every 35 days or seven days, so it's paying a lower short-term rate.
The common stock holders get all of the income on the bonds purchased with their $100 million. On the $50 million they get the difference between the preferred stock rate and the yield on the bonds. So the common stockholders dividend will go down if the preferred stock yield goes up. It is at risk to the extent that short-term rates might shoot upwards.
We don't think this is going to happen. Most of the portfolio managers who run leveraged closed-end funds don't think that is going to happen, but it's a risk and it's risk that they may want to hedge. Q: Okay. Take us through the cash market solution. A: You could just take the proceeds of the preferred stock issue and invest it in short-term securities. But then you're not getting the benefit of the steep yield curve. You're not getting the "leverage." So that's not a good solution. Another solution might be to come to a dealer and buy some [interest rate] caps directly. Caps on the PSA Index or the [London interbank offered rate]. But if those caps pay off, they'll pay off in taxable income. Q: Because they're not connected to a tax-exempt issue? A: Right. But if you buy a variable-rate municipal [security] where the interest rates will go up if the PSA index gets above a certain level, then you'd be able to get tax-exempt income and not give up all the yield you would give up if you just bought a floating-rate security.
Aaron Gurwitz is wearing a new hat. After five years as senior economist at Goldman Sachs & Co., Gurwitz has moved to the capital markets side of the firm.
Gurwitz is now manager of the firm's new municipal capital markets group, and works to integrate the needs of issuers and investors.
As chief economist, Gurwitz was quite a success. In November 1992, he drew gasps when he told a conference audience that the yield on the Treasury's 30-year bond would drop below 7%.
Gurwitz was right on. In February, the long bond dropped through 7% and just kept dropping.
Then, in a March interview with The Bond Buyer, Gurwitz predicted that rates would continue to drop. And they have. The long bond now yields less than 6%.
What does the prognosticator in Gurwitz think of today's rates? At a conference of institutional investors last month in Manhattan, Gurwitz told the audience that rates would keep dropping "until further notice."
Now that his derivatives hat is firmly in place, Gurwitz frets that the term "derivative" has acquired a bad reputation. "Maybe we should call some of these products variable-rate debt," he says.
At times, the interest rate forecaster and the derivatives salesman are at odds. Take the case of municipal forwards.
Investors are anxious to buy municipal forwards, which essentially are contracts to buy bonds not yet issued at a preset price and yield. If rates decline, however, issuers looking in today's rates will be struck with selling the future issues at above-market rates.
"It's difficult to convince issuers to do those trades because our view is that interest rates will continue to decline," says Gurwitz. "And if that's the case, then locking in a forward yield, locking in a refunding yield right now at a higher yield than the spot market, the forward rate, just doesn't look compelling."
In an interview last week with staff reporter Aaron Pressman, Goldman's municipal capital markets guru discussed the use of derivatives as an alternative to cash market trades.
