(bonds down, stocks down -- what's going on?)
Beacon Hill Disaster: The First of Many?
By Aaron L. Task
Senior Writer
10/22/2002 03:44 PM EDT
Revelations of startling losses at Beacon Hill Asset Management are not terribly startling to denizens of the fixed-income world. Over the years, extreme movements in interest rates have invariably caused some bond managers to get caught wrong-footed. The names change, but innocent investors are rarely protected.
Indirectly, the experience at Beacon Hill, which last week announced its two biggest funds had lost more than 50%, or $400 million, speaks to concerns raised in recent weeks about Fannie Mae (FNM:NYSE - news - commentary - research - analysis) and J.P. Morgan (JPM:NYSE - news - commentary - research - analysis). Namely, that big bets on interest rates can backfire with disastrous results.
Concerns about both Fannie and J.P. Morgan have been ameliorated by recent market action, but critics contend they have become leveraged bets on interest rates, and thus subject to the vagaries of that market. This story isn't about Fannie or J.P. Morgan specifically and isn't intended to predict similar results, but seeks to serve as a real-world example of what their critics are concerned about.
Double or Nothing
On the surface, it appears Beacon Hill was undone by heavy refinancing activity and a daring trading strategy it engaged to counteract the impact. In early October, when Beacon Hill reported losses of 25%, or $185 million, it cited "unprecedented, accelerated mortgage prepayments triggered by historically low interest rates."
When mortgages are refinanced, owners of mortgage-backed securities are refunded their original cash investments, which they then seek to reinvest, usually at lower rates. This is know as "reinvestment risk" in bond circles and is one of the biggest challenges to fixed-income participants.
An influx of cash due to heavy refinancing activity sparked the big widening in Fannie Mae's duration gap in August, although the level narrowed in September. To offset the impact of heavy refinancing activity, Fannie Mae was believed to have been an active and aggressive buyer of 10-year Treasury equivalents and other long-dated securities, a common strategy among mortgage-backed participants in a time of heavy refinancing. (The concern of critics such as Jim Bianco of Bianco Research was that Fannie Mae is so big that its buying was having an outsized effect on Treasury yields.)
But unlike Fannie Mae and countless others, Beacon Hill was reportedly shorting Treasuries at the same time it was facing stress in its mortgage-backed portfolio. According to several bond market participants, Beacon Hill was short Treasuries through September and then covered just before prices on the 10-year peaked on Oct. 9.
Apparently, the firm was positioned to benefit from a rise in bond yields and resulting decline in refinancing activity, which would aid its mortgage-backed portfolio. A spokesman for Asset Alliance, a $4 billion New York investment management firm that has a 50% equity stake in Beacon, said it was unlikely representatives of the firm would discuss trading strategies.
In effect, Beacon Hill was making a double-or-nothing gamble and "was basically betting against volatility," said Anthony Crescenzi, chief bond market at Miller Tabak and a RealMoney.com contributor, who said the fund's experience "reminds me of 1998 when risk aversion increased and Treasuries outperformed everything else."
Just as Long Term Capital infamously imploded in 1998 by betting spreads between various corporate and emerging market bonds would return to normal levels vs. Treasuries, so was Beacon Hill apparently betting that extreme declines in interest rates this year, and resulting record refinancing activity, would revert to the mean.
"When volatility picks up and rates move sharply, the [normal] correlation breaks down in a big way," he said. "In this environment, the trade that would hurt the most is being short Treasuries and long mortgages," which appears to describe Beacon Hill's strategy.
But Michael Reiger, a vice president at HedgeFund.net, suggested Beacon Hill likely had a "concentrated bet" on volatility in mortgage-backed securities via a derivatives swap contract.
"Losses of 50% is unimaginable off something as simple as an ineffective hedge against interest rates," Reiger suggested. "Even Long Term Capital didn't lose that much in as short a time frame and they were many times more levered than what Beacon Hill advertised."
The hedge fund observer believes there was something beyond a "simple hedge" because Beacon Hill had reported to HedgeFund.net that it was up 0.72% in the month of August and up 9.9% year to date heading into September. (Then again, Beacon Hill upped the size of its losses as of Sept. 30 to 54% from 25% originally, so maybe those numbers, and the amount of leverage it "advertised," were inaccurate as well.)
"Beacon Hill blew themselves out and you have the rest of the mortgage-arbitrage community humming along," Reiger said. "This seems fairly isolated," which is a view Crescenzi and others also expressed.
Isolated or not, it's unclear what impact, if any, the news of Beacon Hill's losses and Friday's auction of about $275 million of its bonds is having. Treasuries continued to swoon on Monday, and couldn't manage much forward progress midday Tuesday despite sizable losses in stock proxies, sparked by weaker-than-expected results at Texas Instruments (TXN:NYSE - news - commentary - research - analysis).
There's also the issue of Beacon Hill's portfolio of collateralized debt obligation (CDO) which RealMoney Pro.com contributor Brian Reynolds observed might need to be unwound or shifted to another manager if Beacon Hill is "no longer viable."
Maverick or Leader?
Going forward, the question is whether Beacon Hill will be remembered as an isolated incident or the first of a string of losses among fixed-income managers, as occurred in 1994.
As with the past year, 1993 saw a huge rally in bond prices and an accompanying steep increase in refinancing activity. When the bond market rally stalled in late 1993 and refinancing activity declined, it precipitated a series of losses in 1994, including those suffered by hedge funds such as Askin Capital and Vairocana Ltd. That same year, General Electric (GE:NYSE - news - commentary - research - analysis) sold its Kidder Peabody unit to PaineWebber (itself sold to UBS in 2000) after bond trading scandals and steep losses in the CDO market.
According to Crescenzi, fixed-income participants have learned their lesson from the 1994 and 1998 experiences and that "systemic issues will be nonexistent" this time around.
Citing the recent spike in defensive put buying of 10-year Treasury futures, even before the market reversed, he observed that "mortgage-backed holders were buying Treasuries as a hedge against falling interest rates but at the same time were buying [puts as] protection for a 'just in case scenario' for the possibility rates might turnaround sharply."
Crescenzi believes this "hedging of hedges" will diminish the negative market impact of a rise in rates that he concedes has the potential to be "massive" given the decline in yields to 42-year lows.
Investors who've put billions into bond funds in the past two years better hope he's right.
thestreet.com |
