Research & Analysis Home

Trend Following trades during a Flattening Yield Curve

February 7, 2005

A glance at the current positions across a few different long term, multi-market trend following systems today shows some interesting pairs of positions. Andromeda holds a long (looking for prices to go up) position in 30 yr bonds while holding a short (looking for prices to go down) position in Euro Dollars(not dollars at all, but an interest rate derivative tied to the the London InterBank rate, thus similar to futures on the US fed funds rate); Aberration holds shorts in the two year notes and euro dollars while holding long the Euro Bund, and both Fusion and Synergy hold long Euro Bund positions and short Two Year positions at the same time.

What's going on here? How can a trend following system be long bonds and short bonds at the same time? And why on earth would any trend following system be long bonds when the FED is raising interest rates. (As bond rates go up, bond prices go down) Obviously the systems don't adhere to the old market axiom - don't fight the FED.

Being long and short bonds at the same time, and 'fighting the fed' seem counterintuitive, as bonds are usually referred to as a single investment and we hear about interest rates either moving up or down as if there was only one interest rate. But the truth of the matter is there are many different types of bonds, usually categorized by their duration (or length of investment) and just as many different interest rates. System developer know this, and have bonds of different lengths and types in their portfolios.

So despite the fact we have been hearing on the news and reading in the paper for the better part of the last year that interest rates are moving higher, not all interest rates are moving higher (thus not all bond prices moving lower). There are different interest rates for each bond duration (we'll keep it simple by only talking about US bonds here), so depending on whether you're talking about 6 month treasury bills (bills = bonds with duration under 1 year), 2 or 5 year notes (notes = bonds with durations under 15 years), or 30 yr bonds - not only will there be a different interest rate for each, but that interest rate could be moving quite independently from the others.

It makes sense that bonds with different holding periods should have different interest rates. Ordinarily, short-term bonds have lower interest rates to reflect the fact that an investor's money is at risk for a shorter amount of time. No one knows what is going to be happening in say, 30 years, so the interest rate of a 30 yr bond should be higher to reward the investor for taking on more risk.

Economists use a nifty tool called a yield curve to view the different interest rates across the different lengths of bonds (durations). What is a yield curve? Its just a fancy name for a chart showing the interest rates at different bond maturities. Still a little too much economist talk - how about: its a chart that shows the interest rates on bonds spanning holding periods of 1 month to 20+ years on one chart. An example is below:

So we now understand that different bonds have different interest rates. But shouldn't the rates all be moving in the same direction? It seems that the only difference should be in the magnitude of the moves, meaning you would see something like the rate on 5 yr notes rising, but rates on 30yr bonds rising twice as fast.

This is where things get tricky. The different rates of the bonds not only differ in the magnitude of their moves, but can also differ in the direction of the move. Economists call such times flattening or steepening of the yield curve. The charts below show that the past year has seen long term bond prices rise while short term bond prices have fallen. In coming out of a normal yield curve where short term rates were lower than long term rates, this movement is a flattening of the yield curve. A perfectly flat yield curve would be one where all bond durations have the same interest rate, so anytime rates are converging, the yield curve is flattening.

A flattening yield curve where the 2yr note's rate has risen from ~1.7% to ~3.2% (prices down) while the 30yr bond's rate has fallen from ~5.0% to ~4.5% (prices up) has been the story for the better part of the past year and is the answer to why some trend following systems are currently long some bonds while short others.

Sector risk:

A major flaw with the global portfolios of many trend following systems is the risk of severe sector concentration. Nowhere is this more of a concern than in the interest rate sector, where Aberration's global portfolio contains no less than eight different interest rate futures: EuroDollars, 2yr notes, 5yr notes, 10yr notes, muni bonds, Euro bonds, long gilts (UK long term bonds), and Aussie bonds. For comparison, the portfolio contains an average of four and no more than seven markets in any other sector (bonds - 8, currencies - 7, energies - 5, metals - 4, softs - 4, grains - 3, meats - 1, index -1) Other trend following systems have a heavy dose of interest rate futures too, with Andromeda utilizing six in its global portfolio and Axiom long term having five.

The risk of having such heavy sector concentration is that you then have heavy sector exposure. Attain helps investors reduce the risks of sector concentration in long term trend following systems by overlaying portfolios with maximum sector limits. This is commonly misunderstood to mean no more than a certain number of bond markets in any one portfolio, but in fact concerns net positions within that sector.

Investors looking to mitigate sector concentration risk by eliminating sector components from the portfolio are short sighted in that the components of any one sector will not always move in the same direction. We are seeing this in real time in the current flattening yield curve environment where short term bond prices have been going down at the same time long term bond prices have been going up. Sector concentration risk is prevalent when you have many positions on in the same market sector, in the same direction.

In contrast, if you have six bond positions on, but three are long and three are short, your net exposure to that sector is close to zero, as sector wide shocks up or down (the shocks are what you're worried about with too much sector concentration) would result in large gains or losses on the long side being offset by large gains or losses on the short side. For this reason, Attain utilizes a net sector position filter, which can limit market sector concentration risk while allowing for holding larger than normal concentrations in any one sector when the positions offset one another.

- Jeff Malec

IMPORTANT RISK DISCLOSURE
Futures based investments are often complex and can carry the risk of substantial losses. They are intended for sophisticated investors and are not suitable for everyone. The ability to withstand losses and to adhere to a particular trading program in spite of trading losses are material points which can adversely affect investor returns.

Feature | Week In Review | Chart of the Week |

Chart of the Week : The flattening yield curve

Feature | Week In Review | Chart of the Week |

The Week in Review: Fighting the Fed proves profitable for Mesa Bonds & Notes

Don't fight the Fed is an old market saying, but those who have bought 10yr and 30yr bonds while the Fed has been raising rates have been well rewarded. Along these lines, Mesa Bonds has come nearly full circle after coming within a breath of stop trade status in May of '04.

While the system has not yet made it out of the woods in regards to drawdown, it has made an impressive comeback since hitting all time lows in May of â€™04. The system received a little extra boost last week as another weak US unemployment report helped bonds rally +1.09%, increasing the system's open trade equity on its long trade to +$3418.75 per contract. Mesa Notes, meanwhile, is also holding long for open trade profits of $609.25 . So much for fighting the Fed.

Elsewhere, the SP systems which were roughed up in January have hit the comeback trail somewhat. While last weekâ€™s gains were not huge, they did breath some life into a few systems that had a very tough month of January. Overall, the US Stock indices enjoyed their best week of â€™05, as SP futures climbed +2.33% and Nasdaq futures climbed +1.62%.

**Day Trading**

The day trading systems started February off on the right foot as a handful of systems took profits in the first week of the month. Compass SP was the top performer, making +$1112.50 per contract after going long on Friday afternoon. The system was able to catch the majority of the sessionâ€™s late afternoon rally in its only trade of the week. Compass's sister system Clipper eRL also saw success, making $131 per emini Russell.

AG Xtreme and R-Mesa 5 were other systems that benefited from Fridayâ€™s rally, netting +$1075.00 and +$250.00 respectively last week.

Other systems that saw success included the RC systems, with RC Success ES making +$310.00 and RC Miracles ES making +$220.00 per e-mini contract, The Blue Wave systems saw mixed action, with BWT Zones SP making +$85.00 per full size SP contract, but BWT Rockâ€™nâ€™Russ e-RL losing -$151.00 per contract and BWT Zones e-RL losing -$681.50. After factoring in commission costs, Helix SP broke exactly even for the week after trading six times.

Other systems that finished in the red included Daybreaker SP which lost -$317.50 per contract, and Trading Visions, Inc. systems: Impetus e-RL -$31.00 per contract, Spectrum SP -$225.00 per contract, and Spectrum e-RL -$210.00 per contract.

In electronic e-mini trading, Cipher ES lost -$247.50 and Magnitude ES lost -$387.50 per contract, while the Electric Daybreaker system continued to struggle a bit, losing -$297.50 in the ES, -$290.00 in the NQ, -$365.00 in the eRL, and -$140.00 in the eMD.

**Swing Trading**

Swing trading performance was mixed as Axiom and Eclipse held long while Tzar and I-Master held short for the most of the week. The markets favored the long positions, and Axiom ended the week holding open trade profits on the long side of +$1750.00 per contract in the e-MD, +$1510.00 per contract in the e-RL, and +$1120.00 per contract in the e-mini SP. Only Axiom NQ posted negative results for the portfolio losing -$760.00 per e-mini NQ as it reversed twice during the week. .

Eclipse also benefited from the rally as the system is holding long for open trade profits of +$1570.00 per e-mini Russell contract, while Tzar remained short despite the markets up move. The system is losing -$542.50 per e-mini SP contract, -$620.00 per e-mini NQ contract, and -$2060.00 per e-mini Russell contract.

I-Master continues to struggle mightily, as the system posted another losing week across the board. The system lost -$320.00 per NQ, -$1367.50 per ES, -$2080.00 per e-RL, and -$2150.00 per e-MD contract. And finally, Mesa Notes benefited almost as much from Fridayâ€™s bond rally as its sister system Mesa Bonds did. The notes version is holding long for open trade profits of +$575.00 per contract.

**Long Term**

In the traditional commodity markets, Coffee was the big story after rallying 4.94% for the week. Most trend following systems stay away form coffee because of its often violent swings one way or another in a mainly commercial driven market; but Aztec was holding long.

Other markets on the move included foreign currencies, which continue to lose ground to the US dollar. The week saw Eurocurrency fall -1.28% and the Dollar Index gain +1.09%. Andromeda took the dollar rally as a signal to exit the Yen, making $3,418.75 total on the trade over two legs.

Finally, grains were on the move again to the downside, led by soybeans which dropped -2.82% for the week. Andromeda and Brix are holding short in the beans. .

Please Login to: http://www.attainaccess.com for the latest updated statistics.

Feature | Week In Review | Chart of the Week |