August 13, 2007
Hardly a Bailout
The Federal Reserve did exactly what it was supposed to do on Friday.
As I've noted before, under most conditions, the Federal Reserve is irrelevant in the sense that (since the early 1990's when reserve requirements were removed on all but demand deposits) there is no longer a link between bank reserves and the volume of lending in the banking system. However, the Fed certainly has a role to play during bank runs and other crises where the demand for the monetary base soars.
That's exactly what the Fed did on Friday. Contrary to the apparent belief of investors, the Fed did not shift its policy, nor did it “bail out” the mortgage-backed securities market by “buying” them from banks.
What actually happened is that the Federal Funds rate shot to about 6% on Friday morning, and the FOMC brought it down to its target rate by entering into 3-day repurchase agreements . The banks sold securities to the Fed on Friday, and are obligated to buy them back from the Fed on Monday at the sale price, plus interest. Such open market operations are designed to ease the immediate demand for liquidity, and to give the banks and dealers more time to find buyers in the open market for the securities they are trying to liquidate.
This was not a major policy shift. Again, it was an effort to keep the Federal Funds rate at the current target of 5.25%, in the face of demand for base money that was pushing the Fed Funds rate to 6%.
These repurchase (RP) agreements fall into three increasingly broad “tranches:” 1) Treasury securities, then 2) federal agency debt, and finally 3) mortgage backed securities issued or fully guaranteed by federal agencies. “Today's RPs were of this type,” noted The Federal Reserve Bank of New York , which conducts the Fed's open market operations. So the Fed was not taking in the toxic, leveraged, exotic stuff.
Economist Steven Cecchetti concurs, “A quick look at the history of these temporary open market operations shows that they have been taking mortgage-backed securities as collateral for repo for some time. The quantities have normally been small (between $100 mil and $2 bil) but they have been doing it. So this is not what I would call an ‘intervention in the mortgage-backed securities market.' And it is not unusual.”
Now, the size of the operation ($38 billion) was unusual, as was the scale with which the Fed allowed dealers to submit mortgage-backed securities as collateral, rather than simply Treasury and agency securities. My impression is that in doing so, the Fed had no intent of “bailing out” the mortgage backed market, or of creating a huge “moral hazard” by absorbing losses for the irresponsible behavior of lenders. Rather, the Fed had to allow submission of mortgage-backed securities because that's what the banks actually own, and it's precisely the collateral for which the banks can't find a buyer.
Look at Treasury bill yields – they're dropping sharply again because investors are scrambling for default-free securities as a safe haven. Banks and dealers have no problem selling those puppies on the open market, so there's no reason to enter a Fed repo to do it. But banks have drawers full of the mortgage-backed stuff that they can't get rid of, so the Fed bought them more time by allowing them to post those securities as collateral for 3 days. Most likely, the Fed will have to do it again on Monday, but in any event, these are not securities that are going into the “investments” column of the Fed's balance sheet. They are simply collateral taken for short-term credit extended. The Fed does not assume a risk of loss unless the bank defaults on the repurchase agreement with the Fed (whether the mortgages underlying the collateral go belly up is of secondary importance, because it is relevant only if the bank is already in default, and at that point, believe me, we've got bigger problems).
A few interesting details – in the midst of Friday morning's panic, banks would have liked to have done more. At the 8:25 AM operation, $31 billion of securities were submitted by the banks for repo, and $19 billion were accepted by the Fed. At 10:55 AM, $41 billion were submitted, and just $16 billion were accepted. But by 1:50 PM, the scramble for funds had eased somewhat - $11 billion were submitted, and $3 billion were accepted.
Given that about $1.4 trillion of interest-only adjustable-rate mortgages were issued in 2005 and 2006, and hundreds of billions in sub-prime mortgages are already delinquent, a $38 billion repurchase operation by the Fed, where the securities posted as collateral have to be bought back by the banks unless the banks default, is hardly a “rescue operation.”
The Fed has an interest in stabilizing the banking system and the real economy. It has no interest in taking the private sector's loss for the irresponsible lending practices of recent years, nor in saving overly aggressive hedge funds from the losses on their leveraged bets. Again, the Fed did exactly what it was supposed to do on Friday. There will inevitably be enormous losses taken as a result of mortgage defaults – but don't assume it will be the Fed that takes them.
Hedge funds, basis risk, and dispersion
Given the intensified focus on hedge fund losses in recent days, it's important to understand what causes a “hedge fund blowup,” and what distinguishes various approaches.
I used to tell my students at the University of Michigan that there are three factors at work in most financial debacles.
1) Mismatch: The investment position typically involves either a position that relies on a market moving only in one direction, or it involves a combination of long and short positions where there is no natural “offsetting” relationship, or that relationship is purely statistical and susceptible to breaking down under market stress.
2) Leverage: It's hard to really blow yourself up unless you're doing it big, usually with borrowed money.
3) Lack of disclosure and transparency: Generally speaking, blowups are more likely in situations where there is no regular reporting of investment positions, or where very arcane and complex instruments are being used.
A few examples. When Long-Term Capital Management blew up, they had a huge book of long and short positions in international debt securities that were “statistically” related in terms of how they typically behaved. LTCM then took that position and leveraged it up 40-to-1. In other words, for every $1 of capital, they were controlling $40 of securities. Finally, their position (as is typically the case with hedge funds) was a black box, so the hedge fund investors were operating completely on the basis of returns information, with little idea of what was driving those returns. Unfortunately, the fine-tuned “correlation matrix” between their long and short positions blew up in a global debt crisis. At 40-to-1 leverage, it only takes a 2.5% movement between your long positions and your short positions to wipe out 100% of your equity.
When Nick Leeson took down Barings Bank of England , he did it by taking a heavily bullish position in Nikkei futures. As his losses grew, he continued to add to his position, and built up massive leverage. But in order to conceal those losses, he disabled the normal reporting channels by which Barings would have been able to oversee his trading. The combination of leverage, mismatch, and lack of transparency allowed one trader to bring down one of the oldest banks in England .
When Robert Citron drove Orange County into the red, he did it by taking a large position in highly leveraged and difficult to understand “inverse floaters.” These exotic securities essentially wrapped up the three ingredients for debacle – mismatch, leverage and lack of transparency – into a single package. A lot of the more toxic CDO mortgage securities today are of this variety.
In short, when you observe the really big wipeouts, you'll notice that they tend to be in vehicles that take leveraged, mismatched positions and don't provide much disclosure.
A mixed bag of “long-short” strategies
It's notable that in recent weeks, a significant number of funds in Morningstar's “long-short” category have experienced abrupt losses. As Warren Buffett famously said, “it's only when the tide goes out that you learn who's been swimming naked.”
Hedged strategies in the mutual fund area differ from standard “hedge funds” primarily because they use less leverage, or none at all, because they have regular disclosure and reporting requirements, because the mutual fund industry is highly regulated, and because hedged mutual funds don't charge 2% plus 20% of profits (though even the mutuals in this category generally sport much higher expense ratios than HSGFX).
Still, various strategies differ in the amount of “mismatch” they can produce, and the amount of leverage they can take (and I don't know any that do quite as much disclosure as I do in these weekly comments).
There are a number of funds that follow a “130/30” strategy, meaning that they are 130% invested in a portfolio of long securities, and simultaneously short 30% in stocks they view as unattractive. While this structure has greater flexibility, it also runs the risk that the short positions may have no “overlap” with the longs, and may be in entirely different industries. Without that overlap, the worst-case risk is the sum of the absolute exposures, in which case a fund might find itself behaving as if it is 60% leveraged. The risk is much more controlled if the long and short positions represent “pairs” in the same industry, for example, being long GM and short Ford. There are also a few “statistical arbitrage” funds can take significant “basis risk” by having long and short positions. A few of these have had tremendous difficulty lately – I would suspect because there may not be a tight enough overlap of characteristics between the long positions and the short positions.
Of course, what is most relevant to shareholders is how the Strategic Growth Fund is positioned. It's important to note that I wrote controls into the prospectus specifically to limit the ability of the Fund to take leverage or mismatched positions.
While there is no regulatory reason the Fund could not use outright leverage to buy stocks, the prospectus allows leverage only though the use of a small percentage of funds in call options. The reason is that I don't think it's appropriate for a mutual fund to take a position that invests more capital than it has, and relies on the market moving in one direction or another, at the risk of unacceptable losses otherwise. With call options, the most you stand to lose is the premium paid for the option. So while the Fund can allocate a few percent to calls, that few percent is the only potential difference from a fully invested position. There is no potential for a major loss on account of leverage.
Similarly, the dollar value of our shorts never materially exceeds our long holdings – there are certainly “bear funds” that take 100% or even “ultra-bear” 200% short positions in the market. But again, I wrote the restriction against material short positions into the prospectus. The Fund does have the ability to stagger its strike prices somewhat, which may produce modest gains on market declines, and give back part of those gains if the market rebounds quickly before we have a good opportunity to reset our strike prices. But the Fund doesn't establish positions where the “notional value” of our long put/short call combinations materially exceeds the value of our long positions.
Second, I generally maintain a significant overlap between our long positions and our hedges. Now, that's not a perfect overlap – if the Fund was to hold, for example, every stock in the S&P 500 with the same weights as in the index, and was to fully hedge that portfolio with an offsetting position in futures or option combinations, the position would have no risk at all, and (because of the way that futures and options are priced) would earn an interest rate of roughly the 3-month Treasury bill yield. This makes sense – a risk-free position should earn the risk-free rate.
In our actual investment positions, the Fund holds a diversified portfolio typically of 100 or more individual stocks in a wide range of industries, and hedges the market risk of those stocks with offsetting short positions in the S&P 500 and Russell 2000 indices. Note that we don't take our long positions in one set of industries and our short positions entirely in another – rather the short positions are in broad indices that contain the same industries and range of capitalizations, if not exactly the same stocks.
Our largest sources of “mismatch” or what is commonly called “basis risk” is in those industries where we have a much different weighting than the major indices. Currently, for example, our largest weighting difference is in financials, where the Fund has virtually no weight, and the S&P 500 has about 25% weight. Meanwhile, the Fund has about 25% more weight than the S&P 500 allocated to consumer, health care, and technology stocks. We also have a more subtle source of basis risk in the fact that we prefer companies with greater stability and lower debt, so there's a “quality” bias in our portfolio here.
All of this means that the Fund will tend to perform best on days when financials and low-quality speculative stocks are weak, while health care, technology and consumer stocks are holding up relatively better. In contrast, the Fund can be expected to experience a pullback on days when financials and low-quality speculative stocks are strong, while health care, technology and consumer stocks are lagging those sectors.
As always, the basis risk that we accept in the Fund is very intentional – our various weightings in individual stocks and industries is based on specific characteristics of the stocks we hold – valuations, balance sheet stability, price/volume behavior, sponsorship, earnings expectations, and other factors.
The Fund is currently positioned in a way that I expect to achieve gains regardless of the direction of any sustained market movement. We may achieve higher returns to the extent that our stocks perform better than the indices we use to hedge, and may experience losses to the extent that our stocks significantly underperform those indices. Our long-term experience is that the performance increment of our stocks over the major indices has been quite positive, on average. On smaller market movements, the “staggered strike” position of our current hedge is likely to produce a small positive bias on down days and a small negative bias on up days, but that behavior is “local” – over time, the cost of the additional protection from higher-strike put options is essentially financed by the implied interest earned on the hedge.
Needless to say, I don't think it's useful to parse day-to-day fluctuations too finely. It's absolutely reasonable to expect my investment decisions to achieve strong total returns in the Fund over the full market cycle, with less downside risk than the S&P 500 – that is my job, and I readily submit to being evaluated on that basis. The most recent full market cycle runs from the peak of the 2000 bull market to the most recent market peak. Of course, we can also measure a market cycle from one bear market trough to the next. Measured from the 2002 lows, it appears that the trough-to-trough market cycle may have completed its bull phase, but as always, there is no need to forecast – we'll take our evidence as it arrives.
As of last week, the Market Climate for stocks remained characterized by unfavorable valuations and unfavorable market action. While we had lifted a portion of our short calls in the prior week as the market declined into the low 1400's, last week's mid-week market advance was accompanied by tepid market internals. Despite the moderate net gains in the major indices for the week as a whole, declining stocks led advancing stocks. The Fund reestablished a fully hedged stance during Wednesday's market strength, as the conditions we require to identify a more sustained “clearing rally” failed to emerge. As of last week, the Strategic Growth Fund was fully hedged.
Following Wednesday's unusual market action, I posted the following note on the Fund's home page, to help in interpreting that day's returns: “A note regarding HSGFX performance drivers for 8/8/07 - the Fund's 0.25 decline today was attributable to an unusually wide spread between the performance of the stocks held by the Fund and the indices used to hedge. The Fund's stock holdings appreciated by approximately 0.4%, versus a gain of 1.4% in the S&P 500 (led by homebuilders, cyclicals and financials in which the Fund has little exposure), and a gain of 2.7% in the Russell 2000. No individual stock in the Fund was a primary driver. The largest gain in any individual stock holding was approximately $3.5 million, the largest individual loss was -$2.3 million (each representing about 1-2 cents of NAV). The most unusual features of the day were the dispersion of returns between various groups of stocks and the intense focus of investors on battered sectors.”
I noted a few weeks ago that increasing volatility at 5-10 minute intervals tends to be a precursor to significant market weakness. Indeed, a variety of systems, both in the physical world, and in networks, display a “signature” prior to chaotic instability, similar to how tremors precede earthquakes. These signatures are sometimes measured in terms of very arcane features, but in the stock market, you can observe it prior to other historical panics and crashes as a combination of surging trading volume coupled with rising volatility at increasingly short frequencies (so that the time dimension “collapses,” and you begin to observe fluctuations in 10-minute, 1-hour and 1-day periods that would normally take several days, weeks, or even months).
It is of some concern that we are seeing this sort of behavior here, but this does not mean that a crash or further panic should be expected. There are certainly some positive factors, or at least factors that investors believe are positive. For example, Treasury yields are falling (albeit because Treasuries are being sought as safe havens), and stocks look cheap to investors who believe in misleading measures like forward operating earnings, and aren't aware that the historical norms they are applying are actually based on trailing net earnings and normal profit margins. While I believe that these perceived “positives” are largely empty arguments in terms of valuation, we have to allow for the fact that enough investors believe them to potentially act on them.
The bottom line is that at present, both valuations and market action remain unfavorable, and the Fund remains fully hedged. Still, if investors develop enough willingness to accept risk to produce an improvement in market internals, I would expect that we will remove some portion of our hedges. Presently, the case goes to the prevailing evidence, not to any speculation about how the evidence might change, so for now, the Strategic Growth Fund remains defensive.
In bonds, valuations deteriorated but market action, at least in Treasuries, improved. China remains a wild card, and there remains some risk that U.S. trade and currency sanctions (such as tariffs, etc) could provoke some liquidation of Treasuries by China as a countering move. In any event, historically, it has been best to focus first on yield levels and second on yield trends and other market action when setting the duration of a bond portfolio. For now, the Strategic Total Return Fund continues to carry a duration of about 2 years, mostly in TIPS, with about 15% of assets in precious metals shares.
Prospectuses for the Hussman Strategic Growth Fund and the Hussman Strategic Total Return Fund, as well as Fund reports and other information, are available by clicking "The Funds" menu button from any page of this website.
|Home | The Funds | Open an Account | Account Access | Research & Insight | Site Map|
For more information about investing in the Hussman Funds, please call us at
513-587-3440 outside the United States
Site and site contents © copyright 2003 Hussman Funds. Brief quotations including attribution and a direct link to this site (www.hussmanfunds.com) are authorized. All other rights reserved and actively enforced. Extensive or unattributed reproduction of text or research findings are violations of copyright law.
Site design by 1WebsiteDesigners.