Liquidity is the ability to sell a security without significantly affecting its price. This is a positive attribute, and all things being equal, traders will pay more for securities with higher liquidity. U.S. Treasury bills, notes and bonds are among the most liquid instruments in the world, but even they are subject to the effects of liquidity, due to the way they are sold.
Treasury debt is sold at periodic auctions. T-Bills are auctioned weekly (on Mondays), 2 and 5 year notes monthly, and 10 year notes in Feb, Mar, May, Jun, Aug, Sept, Nov, and Dec. “On-the-run” securities are ones being sold at the immediately next auction; “off-the-run” securities are to be sold at subsequent auctions. In the secondary markets, on-the-run securities are always considered the most liquid and thus the most desirable. This is an artifact of cyclical increased demand around the time of a Treasury auction. Since on-the-run securities are more desirable, they trade at higher prices and lower yields than off-the-run securities. The yield difference between the on- and off-the-run securities is known as a liquidity spread.
A liquidity spread trade is one in which the relatively expensive on-the-run securities are shorted and an equivalent face value of cheaper off-the-run securities are purchased. The spread is captured by the trader in exchange for the liquidity risk he/she takes on. This liquidity risk is due to the possibility that in a down market, the less liquid security will undergo a larger price decline than the on-the-run security. Historically, the liquidity spread has averaged 15 to 20 basis points.
Another reason for the relatively higher demand for on-the-run bonds is their use to hedge positions in the when-issued (i.e. a forward contract for an about-to-be-issued security) market. Speculators short the when-issued forward contract in hopes of buying it back at a lower price in the auction. Owners can then borrow cash at a low special repo rate using the on-the-run bonds as collateral. This special rate can be much lower than the corresponding general repo rate, providing an opportunity for riskless arbitrage – borrow at the special rate and lend at the general rate. This extra value of on-the-run bonds, termed a convenience yield, reinforces the yield difference between on- and off-the-run bonds and is thus an added inducement for liquidity spread trades.
The difference between the interest rate on three-month Treasury bills versus the three-month London Inter-Bank Offering Rate (LIBOR), is known as the TED spread. The term “TED” arises from the 90 day T-bills and 90 Eurodollar (ED) certificates of deposit. TED-based strategies can be viewed as credit spread trades, pitting highest quality government debt against slightly lower quality AA-rated inter-bank debt.
The TED spread is measured in basis points (bps) where 100 basis points = 1%. For example, if LIBOR is 4.50% and 3-month T-Bills are trading at 4.10%, the TED spread would be 40 bp. The historic range of the spread has hovered between 10 and 50 bps, except during financial panics or downturns.
Recalling our discussion of the Capital Asset Pricing Model (CAPM), the risk-free rate is used as the basis for finding an asset’s or portfolio’s risk-premium. Well, that risk-free rate is none other than the 3-month T-bill rate – the “T” in TED. LIBOR is by definition riskier, because it has more than zero risk that one of the counterparties to an inter-bank load will default. Therefore, when the TED spread increases, investors are more concerned about counterparty risk, and thus demand a higher LIBOR (or equivalently, a lower price on Eurodollar securities) to induce them to assume the perceived extra risk. Conversely, the TED spread decreases when credit conditions are considered benign.
TED is an inter-commodity spread - you can trade the TED spread by pairs-trading T-bills and Eurodollar CD’s, or more likely, the corresponding futures contracts. If a hedge fund manager feels that credit conditions are going to worsen, he goes long the TED spread by shorting ED futures and buying T-Bills futures. The reverse trade, a short spread, favors the optimistic point of view regarding credit conditions, causing the spread to decrease.
You can also trade so-called term TED spreads, which use longer-maturity (i.e. 6 month, 9 month, etc) securities.
During the financial meltdown of 2008, the TED spread hit a record 465 bps, presaging the collapse of the interbank lending markets. Only massive injections of liquidity from central banks avoided a complete cratering of the financial system. The tendency of such liquidity injections during times of financial crises tends to moderate the long-term volatility of the TED spread.
In our next blog, we’ll examine liquidity-based yield spreads.
Let us resume our tour of hedge fund strategies based on different types of fixed income arbitrage. An asset swap is an over-the-counter agreement between two counterparties to exchange fixed-rate interest payments for floating-rate interest payments. The fixed portion is a position in a fixed-rate loan, such as government or corporate bonds. The floating-rate portion is tied to a reference rate, such as the London Inter Bank Offered Rate (LIBOR) or the Securities Industry and Financial Markets Association (SIFMA) Municipal Swap Rate.
Asset swaps are similar, but not identical to, interest rate swaps (IRS). IRS contracts are based on a notional principal – for instance, a $100 million interest rate swap doesn’t require the counterparties to spend that amount on debt; rather the counterparties are obligated to exchange interest payments as if they owned that much debt. In contrast, an asset swap involves the exchange of interest payments between a real bond position against a floating rate notional position. An asset swaps allows a bond holder to change the cash flow characteristics of its bond position. The long side (the bond holder) of an asset swap pays fixed and receives floating from the short counterparty. In a reverse asset swap, the roles are reversed.
Asset swaps can help hedge different risks, such as credit risk. If I own a risky fixed-rate bond position and swap it out for a floating rate cash flow, I continue to receive that cash flow even if my bonds default. The price of this insurance is that I receive a floating amount based on an interest rate that is discounted from LIBOR. The spot rate for swaps is called the swap rate, and the amount of the discount or premium represented by a swap rate relative to LIBOR is the spread.
Let’s take a concrete example, where the long counterparty has a position in Euro-denominated corporate bonds. We want to figure out the asset swap price (i.e. the spread) on this transaction.
|Issue Date||31 March 2010|
|Maturity Date||31 March 2017|
|Fixed Coupon||5.5% annual|
|Price (With Accrued Interest)||105.3% of par|
|Price (W/O Accrued Interest)||101.2% of par|
|Current Swap Rate||5%|
The long counterparty pays 105.3% of par value to buy its bond position. The long would receive the fixed coupons of 5.5% of par value. The long counterparty then enters into an asset swap with the short counterparty, which happens to be the broker that sold the bond position to the long side. The long pays the fixed 5.5% coupon to the broker and receives a LIBOR-based (LIBOR +/- spread) floating rate cash flow.
To calculate the swap spread, two factors must be considered:
- The excess of the fixed coupons over the market swap rate is paid over the swap lifetime to the long counterparty. In this case, this excess is 5.5% – 5% = 0.5%.
- The present value of the difference between the (“dirty”, meaning with accrued interest) bond price and par value, spread over the swap lifetime. Since in this case the bond trades at a premium to par, the long side must pay this amount to the short counterparty. In this case, we will assume the amounts to a payment of approximately 0.23% for the term of the swap.
These two elements result in a net spread of 0.5% – 0.23% = 0.27%. Therefore the asset swap would be quoted as LIBOR + 0.27% (or LIBOR plus 27 basis points).
Hedge fund managers use asset swaps to tailor their risk exposures. An outright position (long or short) establishes which risks the manager is willing to accept. A boxed position, which is a mixture of long and short positions, allows a manager to hedge his/her risks.
In our next installment, we’ll examine the TED spread.
In preparation for resuming our exploration of hedge fund trading strategies, we took a detour last time out to learn about bond futures and conversion factors, which are used in basis trading. Basis trading is a form of fixed-income arbitrage that seeks to benefit from a change in the spread between a spot bond price and an adjusted futures price. The formula for the basis is:
B = SP – (FP x CF)
B is basis
SP is spot price of bond (clean)
FP is futures contract price (clean)
CF is conversion factor
Clean prices are ones in which the present value of future cash flows, such as interest payments, are not included in the price. Normally, one purchases a bond at the dirty price, which includes such cash flows.
A bond basis trade is the simultaneous purchase and sale of a bond and a bond futures contract to capture a non-zero basis as profit. It is also known as a cash-and-carry trade. (There is an alternative method of achieving the same result using exchange of futures for physicals (EFP) which we’ll not discuss in this article).
The conversion factor is the key to a basis trade. Here’s how it works:
- If the basis is negative, the bond spot price is less than the adjusted futures price. In this case, you would “buy the basis” by buying cash bonds and selling futures contracts. Say that the conversion factor on the 8 ¾ T-Bonds of 5/15/2017 is equal to 1.077. To buy $100M of the basis, you purchase $100M face value of the bonds and simultaneously sell 1,077 (= $100M * (1.077 / $100K) of bond futures.
- If the basis is positive, the bond spot price is greater than the adjusted futures price. Here you would “sell the basis” by selling cash bonds and buying futures contracts. For instance, if the conversion factor on the 7 5/8 T-Bonds of 2/15/2005 equals .0960, then selling $10M of the basis would require selling $10M face of the bonds and buying 96 futures contracts.
To close out the trade, you need to purchase your short position and sell your long position. If you compare the opening spread with the closing one, the difference is the change in basis during the holding period. A narrowing spread favors the short position; a widening spread benefits the long.
If you think the spreads will narrow over time, you benefit from selling the expensive bonds and buying the cheap futures contracts if your prediction is correct. The change in basis over time is the cash part of the trade. The carry portion consists of coupon payments less financing costs (at the repo rate) for the bond (including accrued interest). You realize a profit if the sum of the cash and carry portions are positive.
Every 1/32 of a basis point is worth $31.25; on a position of $10M face, this equals $3,125. Therefore, if a basis narrows by 2.6 basis points, the short position profits on the cash portion of the $10M face trade by 2.6 * $3,125 = $8,125. As long as the short’s carry costs do not exceed $8,125, he/she will pocket a profit.
The risk in a basis trade is that the basis will move in an unfriendly direction due to a change in the yield curve, and/or the repo rate will change to your disadvantage. These changes are important because of the following reasons:
1) If the repo rate decreases, or if the yield curve steepens, carry and basis increases
2) A decrease in the bond’s yield relative to other deliverable bonds will increase the basis
3) Bond duration can affect a bonds response to yield changes: the basis of a low-duration bond will tend to rise with bond yields, whereas the basis of a high-duration bond increases when bond yields fall.
4) Volatility affects technical considerations involving the short’s strategic delivery options. A rise in volatility would tend to lower the futures price and raise bond basis.
There is extensive literature on basis trading, and interesting parties are urged to seek it out before embarking on any trading activity.
Our next topic will be asset swap trades.
Our review of hedge fund trading strategies continues with a discussion of yield-curve arbitrage (YCA), a form of fixed income arbitrage. I have previously written about the yield curve, convexity, and duration. Recall that for bonds not offering embedded features (such as puts and calls), a bond’s price and the interest yield move in contrary directions, giving an inverse association involving duration and yield. Higher yields mean shorter durations. The $duration of a bond is product of the duration and the price (value); with units of dollar-years, it reflects duration changes in dollars rather than in percentages.
A parallel shift in a yield curve occurs when the yield on all maturities change by the same amount. More likely are changes in which the spread between short and long maturities increase (steepen) or decrease (flatten). A dumbbell portfolio is loaded up on bonds at the short and long ends of the yield curve; conversely, a bullet strategy involves the purchase of intermediate-maturity bonds. Yield-curve arbitrage is a trading strategy in which a trader exploits relative mispricings along the yield curve due to high institutional demand for selected maturities, among other reasons.
A well-known form of YCA is the so-called butterfly trade: long dumbbells (the “wings” of the butterfly) and short bullets (the butterfly’s “body’) in a net-zero $duration spread trade. For example, you might set up a portfolio in which you are long 4-year and 8-year maturities, and short 6-year maturities. Small parallel moves in the yield curve would have little effect on this portfolio, since it has a net $duration of zero. However, large parallel moves in either direction will guarantee a positive return due to the positive convexity (yield vs. price) of the portfolio – in effect, one expects greater convexity in the wings than in the body. That sounds good in theory; in practice, yield curves usually experience complex movement patterns that can have an unexpected affect on the outcome of a butterfly trade.
There are four popular types of butterfly trades:
- Cash and $duration neutral weighting – No cash is needed up front, since the cost of the long positions is offset by the proceeds from the short sale. Suitable prime brokerage structures are available such that the long position acts as collateral for the short position, so that zero cash flow is required initially. This strategy benefits from a flattening of the yield curve, because most of the $duration is in the wings of the butterfly.
- Fifty-fifty weighting regression – The trade is structured such that each wing of the butterfly has equal $duration. This strategy benefits from small changes to the yield curve, because the body is less convex than the wings. The position profits from a steepening of the yield curve. Note that this trade is not cash neutral, so return must exceed the cost of carry.
- Regression weighting – A sophisticated trade in which the linear regression measuring the spread between the short wing and the body is regressed against the spread between the body and the long wing. The more-volatile short wing is more likely to move away from the body than is the long wing. So, say for example we determine a regression coefficient of 0.5; it means that a 20 basis-point change in the short wing spread would imply a 10 basis-point move in the long wing spread. Since most of the $duration is in the long wing of the spread, the strategy benefits from a flattening of the yield curve.
- Maturity weighting – The relative maturities of the three components (short wing, body, and long wing) are used as the weighting of each component. The results of this strategy are very similar to the regression weighting scenario, except that the weighting factor will generally be higher than the regression coefficient.
There are calculated risk measures that can be used by traders to determine whether the spread on each of the butterfly strategies is attractive and invites investment. Advanced readers can look up the model developed by Nelson and Siegel to see how to partially hedge the risk exposures of different butterfly spreads.
Next time, we’ll continue our survey of fixed income arbitrage by taking a close look at basis trading.
 Nelson, C.R., and Siegel A.F., (1987) “Parsimonious Modeling of Yield Curves”, Journal of Business,
60 (4), , p.473-489.
Fixed Income Arbitrage (FIA) is the name given to a family of trading strategies that, to various extents, use spread trades on debt instruments to take advantage of pricing inefficiencies independent of overall market direction. A spread trade is the simultaneous purchase and shorting of related securities (and their derivatives) in the hope of profiting from the widening or narrowing of the spread (i.e. the prices) between the two securities. FIA typically deal with large quantities of highly liquid debt instruments, such as government and corporate bonds, asset-backed securities, and debt-related derivatives like swaps, futures, and options. Positions are usually leveraged from 5 to 15 times the asset base’s value, although there is no hard and fast rule concerning this.
Because FIA spreads contain long and short legs, they tend to cancel out systematic market risks, such as changes to the yield curve. Managers are free to hedge away specific risk exposures, including risks due to changes in interest rates, creditworthiness, foreign exchange risks, and default, though the extent of hedging employed varies greatly among hedge fund managers. Managers are also free to take directional positions instead of relying solely on pure neutral hedges.
Spreads available to FIA traders are typically small, which is why leverage is widely employed. Leverage is gained through the use of borrowing, repurchase agreements (repos), and derivatives. It is not unusual to put on an FIA trade for a $100 million notional amount requiring less than $1 million of posted collateral. Usually, the more basic types of FIA trades use higher levels of leverage than do more complicated trades (such as mortgage-backed security trades) that expose positions to particular risks.
FIA returns are made up of spread profits, due to systematic risk premia and/or price inefficiencies, and carry, which is the excess of positive cash flow over negative cash flow. A simple example of carry would be a long position that earned 5.25% interest paired to a short position paying out 5.05%; a 20 basis point carry profit can be achieved by this spread.
Returns from FIA trades can result from sudden market dislocations, demand or supply shocks, changes in investor preferences, restrictions on particular instruments or markets, credit rating changes, execution of options embedded within debt securities, and any event that changes a bond’s anticipated cash flow. Price inefficiencies can arise from a number of factors:
1) Agency bias: the tendency of fiduciaries to purchase securities based on past results
2) Structural impediments: the trading of securities for non-economic reasons relating to tax, regulatory, and accounting issues
3) Market segmentation: the preference for a particular range of maturities can become pronounced enough to cause price dislocations between different securities
Systematic risk premia can result from several causes. For instance, a hedged spread may feature long and short positions that differ in liquidity or credit quality. In this case, a premium is earned by holding lower quality or less liquid positions and shorting higher quality/more liquid positions. Other systematic risks can arise from, among others, cross-currency trades, and spreads between Treasury and agency debt.
In general, FIA traders earn their premia by taking positions that profit when “the sky doesn’t fall”. This is similar to selling a disaster put – the buyer would profit only in the case of catastrophe, in which case the buyer can put securities to the seller at a pre-catastrophe price. This is a form of insurance; you might recall credit default swaps, which served a similar purpose. Normally, it works, and the put seller pockets the put’s premium when it expires without intervening economic disaster.
FIA trades behave much like the short put strategy. When turmoil occurs, there is a flight to quality that causes bond spreads to widen and liquidity to dry up, all of which cause FIA traders to realize losses. Recall that FIA traders want spreads to narrow. This “short volatility” position backfired big-time in 2007-8 during the real-estate crises. The crisis demonstrated the concept of “fat tails”: an event can be so unlikely that its occurrence is predicted only at several standard deviations from the mean, but in fact occurs more often than a normal distribution curve would imply. As a crisis unfolds, market participants all exit at the same time, and leveraged positions collapse. Bankruptcy and financial ruin can easily follow. The collapse of Long Term Capital Management in 1999 is a famous object lesson in this regard.
Now that we have a broad overview of FIA, we’ll begin to delve into particular strategies. Next time: yield-curve arbitrage.
Recall that yield curves (also known as the term structure of interest rates) plot debt maturities (the independent variable) against interest rates (the dependent variable). Debt maturities indicate the length of the borrowing period for a debt instrument. We spoke last time on how a yield curve is shaped; today we’ll look at a few theories that attempt to explain yield curve behavior.
Pure Expectations Theory (PET)
In this theory, it is assumed that any maturity of debt can substitute for any other through the miracle of compounding. For instance, if you have a view as to what the one-year interest rate will be one year from now (the forward rate), then you can determine the current two-year interest rate as the compounded sum of the current one-year rate and the one-year forward.
This generalizes to the geometric mean of short-term yields as the determinant of long-term yields. A geometric mean differs from an average in the it is calculated by taking the nth root of the product of n terms. For example, if you have two terms, say 4 and 16, the average is of course 10, but the geometric mean is the 2nd root – the square root – of their product (64), or 8.
The Pure Expectations Theory accounts for the fact that yields tend to change together over time, but doesn’t explain the fine details of the shape of the yield curve. It posits that forward rates are perfect predictors of future rates, which they are not. It thus ignores interest rate risk and also reinvestment risk. The latter is the risk that one cannot reinvest interest payments at an expected rate. The theory also assumes that the ability to arbitrage among different maturity bonds is minimal.
Liquidity Preference Theory (LPT)
This is a variant of the Pure Expectations Theory. It basically adds a premium to the PET-calculated yield for long-term debt to account for investor preference for short-term bonds over long-term ones. This premium is called the term premium or the liquidity premium. It acknowledges the risks involved in holding long-term debt, which is more likely to experience catastrophic events and price uncertainty than is short-term debt. A second premium is also included in LPT, for default risk, which is more likely when holding a bond for a long period of time, once again due to uncertainty.
Market Segmentation Theory (MST)
This theory acknowledges that different maturities of debt cannot be substituted for each other. This results in separate demand-supply relationships for short-term and long-term debt. Since investors (assumed to be risk-adverse) prefer the less risky short-term maturities, the demand for short-term debt is higher than that for long-term debt, and thus prices of the former are higher, driving down their yields. This helps to explain the normal shape of the yield curve, but not the fact that long and short term rates tend to change in unison, since they are supposed to be two separate and independent markets.
Preferred Habitat Theory (PHT)
Preferred Habitat Theory is an extension of MST which posits maturity preferences, or habitats, for debt investors: some investors like 3-year bonds, some prefer 6-year maturities, etc. If you want to sell an investor a bond outside the investor’s preferred investment horizon, you must offer the investor a premium. Since it is assumed that more investors have short-term habitats, it explains the higher yields on long-term debt, and is consistent with the tendency of short- and long-term debt yield curve segments to retain their shape when overall yields change.
Now that we have a good understanding of yields and the yield curve, we can resume our review of hedge fund strategies. Next time: fixed-income arbitrage.
Loyal readers know that we are currently surveying hedge fund strategies. As we pivot from equity strategies to fixed-income arbitrage, we will first take a short “time out” to learn about yield curves. In this article we’ll discuss yield curve shapes; next time out we will explore the theories that attempt to explain yield curve behavior.
Yield curves (also known as the term structure of interest rates) plot debt maturities (the independent variable) against interest rates (the dependent variable). Debt maturities indicate the length of the borrowing period for a debt instrument. The interest rate associated with a given borrowing period is a point on the yield curve. Curve segments between actual maturities are interpolated. Yield curves are specific for a given currency – the yield curves for U.S. dollar-denominated Treasury bonds will (almost always) differ from the analogous chart of U.K. pound-denominated gilts.
Debt yield is the overall rate of return on a debt instrument. Debt that is “locked up”, as in a certificate of deposit, will usually offer a higher yield than an on-demand savings account, due to the higher certainty of holding the former to maturity. Yield curves normally ascend with time, but the rate of increase diminishes with increased time. The fancy phrase for this is an asymptotically upward slope. It reflects the fact that it is riskier to hold longer maturities, because it’s harder to predict distant future interest rates than it is to predict near term interest rate. This increased uncertainty usually demands a risk premium, i.e. a higher interest rate (known as a liquidity spread). This makes the most sense if investors anticipate a period rising short-term interest rates – current investors willing to tie up their money today must be compensated for forgoing higher interest rates tomorrow. But even if rates are not forecast to rise, the liquidity spread tends to give longer maturities higher yields.
An inverted yield curve is, as you might suspect, one in which short term rates are higher than long term ones. Inverted curves indicate an anticipated drop in interest rates over time. This is not considered a good thing – inverted yield curves are often associated with strong recessions and depressions. Forecasts of a weak economy motivate long-term investors to agree to lower yields, on the premise that yields may go even lower. On the plus side, inverted yield curves indicate a belief in low future inflation. However, in an economic panic, a flight to quality may increase demand for, and thus lower the yield of, long-term government bonds.
Currency is a prime determinant of yield curve shape. Another determinant is the type of debt instrument: government bonds, bank debt, corporate bonds, and asset-backed securities. The latter three also vary by the creditworthiness of the issuer – debt with ratings of Aa/AA and above is less risky than lower-rated debt, and hence demands less of a premium over the interest rates for government debt. The London Interbank Offered Rate (LIBOR) is a benchmark rate reflecting the interest rates top London banks charge each other for unsecured funds. LIBOR is important in the swap market because it is used to peg the floating rate leg of an interest rate swap. For this reason, the LIBOR yield curve is often referred to as the swap curve.
As we noted, the normal yield curve has a positive slope, reflecting investor sentiment that economies will grow over time and that inflation rates will rise accordingly. Central banks, such as the U.S. Federal Reserve, control a country’s money supply in order to either keep a lid on inflation (by tightening the money supply, i.e. raising short term interest rates) or fight recession and deflation (by increasing liquidity and reserves in credit markets, i.e. lowering short term rates and/or printing more money).
Historically, the 20-year Treasury bond has yielded a spread of two additional interest rate points above the three month Treasury bill. However, events can cause yield curves to steepen (larger spreads) or flatten (smaller spreads). Recently, the spread between 2-year and 10-year Treasuries widened to record-setting amounts approaching 3%. Sometimes, a yield curve is bicameral, or humped, meaning that medium-term yields exceed long- and short-term ones. This can occur when short term volatility is expected to be more significant than that in the long term. Flat yield curves can predicts steady interest rates or reflect general uncertainty about the economic future.
Several months ago we began a conceptual review that will eventually lead us to evaluate hedge fund replication. We have so far looked at various pricing models, described systematic (beta) and non-systematic (alpha) returns, and have started to explore different hedge fund trading strategies. We continue today with short-selling strategies.
Shorting a stock means selling shares you do not own in the hopes of benefiting from a decline in the price of the shares. The seller borrows shares, usually from a broker, and delivers them to a buyer in return for cash. The cash proceeds are parked with the lending broker as collateral, where it earns short rebate interest for the seller. At some future point of time, the seller will need to return the borrowed shares to the broker by going out into the open market to purchase replacement shares. Any price decline between the original sale and the subsequent purchase is profit for the seller. Conversely, if the shares appreciate over the interval the seller will register a loss. The capital gain or loss, plus the short rebate interest, minus any transaction costs (i.e. broker fees, stock-lending fees, exchange fees, etc.), represents the net P&L to the short seller.
In addition to holding the seller’s cash proceeds as collateral, most lending brokers require an additional pledge of cash or cash-equivalents (such as Treasury bills) in an amount ranging from 30% to 50% of the shares’ market value. In the United States, SEC rules stipulate that short sellers must already have custody (either through ownership or borrowing) of shares sold.
Some managers execute short-selling strategies without the use of borrowed shares; rather, they utilize derivates to create downside exposure. Forwards, futures, and options (on indices and stocks) are suitable derivatives for a short-selling strategy. The collateral requirements on derivatives are much less onerous, and provide short-sellers with additional leverage on their investments. Other managers hedge their short sales with long positions that partially offset the short ones. This creates a Long/Short strategy with a short bias.
One shorting technique, called shorting against the box, is not part of the hedged equity short selling strategy. Rather, it involves offsetting all of a long position with a short sale in order to lock in a profit on shares that, for various reasons, an investor does not want to sell at the current time.
Long and short hedging strategies share a number of similarities, including careful stock picking and timing. Many practitioners of either strategy engage in bottoms-up fundamental analysis to identify mispriced stocks. However, a short strategy fund manager is also likely to concentrate on certain negative factors, such as aggressive accounting techniques or revised quarterly earnings estimates. But it is not valid to think of a short selling strategy as simply the inverse of a long strategy – here are a few ways short selling has unique risks:
- Stock-borrowing considerations: Short sellers need to locate shares to borrow, which for a hot stock may be expensive or unavailable. They must be prepared for the lender to suddenly call back its shares, and they must negotiate a good short rebate rate.
- Short squeezes: If a stock suddenly gains in price, a short seller may be forced to terminate his/her position prematurely by purchasing high-priced replacement shares, thus locking in a loss.
- Performance/exposure: Long strategies benefit from higher prices not only by higher market values but also because the winning shares now represent a larger portion of a portfolio, and thus increase relative exposure to a profitable stock. The situation is reversed for short sellers – loser stocks do engender a profit, but they lose portfolio share as the price declines, and hence decrease the short-sellers relative exposure to the shares.
- Uptick rule: The U.S. recently reintroduced the uptick rule, which states that shares can only be shorted after an uptick. In practice, hedge funds and other investors are usually able to manufacture an uptick when one is needed, but this is not foolproof.
- Unlimited upside risk: Stocks have a downside limit of zero but theoretically no upside limit. This creates an asymmetrical risk for short sellers that can have a negative psychological impact on the desire to short sell.
These factors, plus certain cultural and/or legal barriers to short selling, including a general reticence by investment analysts to issue “sell” recommendations, can tend to depress short selling activity. Studies show that barriers to short selling artificially inflate stock prices. Thus, short selling can be seen as an effective method to arbitrage “correct” prices in stock markets.
A pure short strategy, much like a pure long one, can benefit from investor skills. What is not clear in either strategy is whether returns are consistently based on superior investor skill (alpha), on systematic exposure to one or more risk premia (beta), or a mixture of the two.
Next time we will start exploring relative-value trading strategies.
We now turn to equity trading strategies that attempt to decouple market movements from portfolio returns. Traders take positions in pairs of similar stocks, longing the “undervalued” one and shorting the “overvalued one”, thereby placing a bet on the ultimate outperformance of the long position. By using equal dollar investments on each member of the trading pair to effect an overall neutral long/short position, the trader is looking to remove beta as a factor affecting portfolio performance. The pair of stocks is usually composed of close competitors in the same market, and therefore presents the same country risk, currency risk, market capitalization risk, etc. By attempting to limit risk exposure to the relative merits of the two stocks, hedged equity market neutral strategy (HEMNS) can be equally attractive in up and down markets.
HEMNS is usually considered a long-term strategy, and may utilize a mix of fundamental and technical analysis. An alternate form of HEMNS is statistical arbitrage (stat arb), in which the pair of positions is determined through the use of technical and quantitative analysis. The goal is to take positions in which a traders proprietary algorithms show a pair of stock to be temporarily mispriced (one too high, one too low); the trader is betting that the stocks will shortly revert to their mean price, at which point the trade is liquidated.
In this example, the trader opens the position with equal dollar amounts of a pair of closely-related auto parts stocks. During the period, each stock pays a dividend. The position is put on for 90 days. Each stock pays a dividend during the period. After accounting for transaction costs, financing costs on the short position, and dividend payments, the strategy provided an annualized return of over 42%. (This is for purposes of illustration only).
For HEMNS to work, there must be a short-term suspension of the efficient market hypothesis (EMH), since a profit opportunity only arises if there is a difference between a stock’s theoretical fair market value and its current price. However, there can be multiple factors at work in HEMNS besides “incorrect” pricing. For instance, style risk arises from choosing assets according to value, capitalization, or momentum. There are also risks due to liquidity, leverage, interest rates, etc. Since each of these risks are rewarded by specific premia, it may be difficult to say for certain that HEMNS returns are due solely or even largely to an asset manager’s ability to differentiate overvalued and undervalued stocks. Such ability, if consistently manifest over time, could be termed alpha, as it may signify returns due to a manager’s skill, not just risk exposure.
However, studies of several paired-equity styles, such as high-value/low value, winning momentum/losing momentum, and small cap/large cap, have shown systematic risk premia available that are not captured by the standard CAPM, as we have discussed in previous articles. Some argue that these risk premia are due to factors like recession risk, bankruptcy risk, and liquidity risk. Others favor behavioral finance, which claims that market inefficiencies are due to suboptimal investor decisions, as an explanation. Investors may not really care, as long as returns are good.
Later on in this series, it will be our task to see if we can replicate HEMNS results without any special stock-picking skill. But we have many more strategies to examine first. We will continue our tour of hedge fund strategies next time with equity hedged short selling.