Doug Kelly is a senior v.p. at Merganser Capital Management in Boston, where he is in charge of the structured finance portfolio. The firm holds roughly $1.4 billion in asset-backed investments.
How is your portfolio allocated?
We have a pretty good representation of most of the major types. Auto, credit card and rate reduction bonds each account for about 25%. Equipment receivables are probably another 10%, in a mix of types including agricultural and construction equipment. And, we have a somewhat smaller commitment to small business and medical office equipment.
What's an investor to learn from the situation with DVI Inc.? [Editor's note: DVI is a medical equipment leasing company that recently filed for bankruptcy. Losses on its asset-backed portfolios have increased since the filing and the issuer has missed a recent payment to ABS investors.]
It's not a simple thing to transfer servicing on that type of collateral. It's not like taking over a bank portfolio of automobile loans where you have a perfected security interest in the collateral. The servicing on those types of assets is really critical. Even where there is a wrap, I tend to be hesitant. We do not own DVI bonds.
Investors are paying more and more attention to the servicers. How does that role factor in to your investment decisions?
A lot of it comes down to how easy it is to transfer the servicing in the event of a problem. Anything where there's unique collateral or unique servicing requirements makes it that much more difficult. DVI is a good example. Another example is Heilig-Meyers Inc., a furniture retailer that had a very atypical servicing methodology whereby consumers were encouraged to go into the local store to pay their balances. If it's a high quality bank credit card portfolio, there's a precedent for being able to take that over. Banks trade their card portfolios without a blip, but when you get into sub-prime, you have to be more on top of the customer, it's a very different type of servicing. When you get away from the non-standard collateral, it becomes more important that the servicer be a going concern and you have some high degree of confidence that they will be around. And, if not, there needs to be a backup servicer.
Home equity issuance has exploded this year. Do you think investors are paying enough attention to the risks in these transactions, or are they letting the low rate environment and steep yield curve mask potential problems for a lot of these issuers?
I think there has been a rationalization in the home equity field, to where a lot of the shakier underwriters are no longer in business. There's less servicing risk, but that's not to say there's not structural risk. If you're buying a new issue sequential home equity loan transaction, assuming the typical ramp up might be to a constant prepayment rate of 23-24%, and if you buy it thinking you are going to get swaps plus 40 or 50 basis points, I question the viability of the assumptions. You're two-year could turn out to be a five-year, and if you paid par for it, the yield is the same. With the curve as steep as it is, you might end up with LIBOR minus 40. You have to be careful about the assumptions you make, because the collateral does have a fair amount of negative convexity and it's not clear you're adequately rewarded. It's hard to evaluate the value of the option--you often have to model your own option-adjusted spread projections, which is not an easy task.
What about manufactured housing? The Federal Home Loan Bank wrote down a big chunk of MH bonds late last month, while at the same time, Warren Buffett seems to be pretty confident in the business line, with Berkshire Hathaway planning a $1.5 billion debt issue to finance its MH line. What's one to take away from all of this?
It's still a sector of the market with a lot of issues. The underwriting has tightened up a lot, but there are still a lot of transactions out there, particularly the late 1990s vintage Conseco Inc. paper, where losses are still in the ramp up period. There's still inventory overhang, a lot of wholesalers are no longer in business. It's not easy to recover losses on repossessions, and if you look at some of the losses on the Conseco transactions, they are probably approaching 80% or north of that. It's possible to find the available interesting bond, but we've scaled back our focus. We do own some Ginnie Mae manufactured housing now involved. It's an important segment of the real estate-related market and there will be some issuers left standing.
What asset classes do you expect to outperform for the rest of the year?
I still think there is value in some of the off-the-run issues in the more straightforward collateral types, such as a regional bank that access the market for the first time. No one knows who they are, but they have a high quality auto portfolio and you might get an extra 10 basis points for the lack of recognition and liquidity. Some of the premium rate reduction paper is still cheap to credit cards. Some of the equipment lease transactions are cheap. And some of the wrapped sub-prime auto transactions are cheap if you can tolerate the headline risk, such as AmeriCredit Corp. Although I did not participate, the wrapped transaction they just came out with was relatively cheap, and it was kind of a food fight.
Any other thoughts on the ABS market?
It's been important to avoid some of the sectors where there's not a lot of diversification in the collateral, or where it's difficult to get your arms around the nature of collateral, such as airline equipment trusts, franchise loans and tobacco loans. These are all things we are staying away from for structural reasons. We try to seek value in structure, in exchange for some liquidity. We usually buy the seniors, but we do have some accounts that permit subordinate or mezzanine investments, and that's one area that I do think is cheap right now. There's been a lot of spread compression in the corporate market and equivalent-rated ABS subs have not followed suit anywhere near to the same degree. While they don't have the liquidity and are often not ERISA-eligible, I think there's some room to tighten in that area.
Do you buy CDOs? Do you have any interest in acting as a collateral manager of your own deal?
We haven't been involved. You're in effect buying into a blind pool where you don't have any control. If we were ever able to partner with someone where it made sense, we would consider it. If someone came to us, it's possible we might act as a collateral manager, but it's not something we're actively pursuing.