How Systematic Investment Management Can Improve Returns

I made the decision this week to switch half of our Income Allocation within portfolios to a systematic management system.  As I mentioned last week, I’m not totally sold that interest rates have bottomed but there’s a strong chance that they have, marking the end of the 35 year bull market in bonds.  If rates have bottomed (meaning bond prices have topped), this will put pressure on bond returns moving forward as interest income is offset by a decline in the price of the bond.  If you have exposure to bonds in your portfolio via a fund, as most people do, it mathematically cannot outperform the current yield to maturity between now and some point in the future if yields are higher at that future point than they are today.  With yields already very low, this means bond fund returns will be even worse, and that just won’t do.

A systematic approach to managing money is rules based (meaning you remove all discretionary decision-making) and quantitatively driven (meaning there is math behind the scenes giving you buy and sell signals).  Most trading systems are momentum based with a focus on being invested for most of the uptrend and getting out when it loses momentum and starts to roll over to avoid the ensuing drop.  This means you never catch the exact top or bottom but hope to catch the 80% in the middle.  The term “invested” is actually a misnomer since this is trading, not investing.  Investing is taking a long-term view and holding an investment through the ups and downs because you’re confident you’ll earn a strong return over time.  It’s more passive and more efficient (expense and tax-wise) than trading, but trading can offer a much better risk/return profile if you’re able to side-step big drawdowns.  I’ve found over the years that most people call themselves investors but actually want the risk/return profile of short- to intermediate-term swing trading.  There’s absolutely nothing wrong with this and I actually believe it can be very beneficial for a portfolio, especially as an investor approaches and enters retirement.

There are tons of trading systems – everything from short-term day trading to long-term asset allocation systems for investors.  Most of these will incorporate calculations for things like moving averages, momentum, volatility, asset class correlations and even economic data to determine the direction of the trend.  Regardless of the data that drives the system, the primary goal is to improve returns by avoiding the large drawdowns.  If the bond bull market has ended and we’re entering a long period of tough sledding, then I want to use a system to help determine the trend so we only own certain bonds when they’re moving higher, and we’re out when moving lower.

If we were looking to build a strong trend following system, here are the key attributes I think we should look for:

• Clear, sustainable trends where we can ride most of the upswing and avoid most of the downswing
• More volatile price action to avoid big drops but be in for bigger bounces
• An asset that has terminal value

Terminal value refers to a bond’s eventual maturity at face value.  Stocks don’t have a maturity date unless they either go bankrupt (not good) or are taken over (usually good).  Since a bond will eventually mature at face value (typically a price of 100 or “par”), unless the issuer defaults, then a drop in the price of a bond below par, to say 80, means you have an added gain baked in the cake in addition to the interest earned.  This means we can rely on bonds to almost always experience a strong bounce following a big dip, especially if prices dip far below 100. THere are some bonds that meet the above criteria:

• High Yield (aka “Junk”) Bonds
• Floating Rate/Senior Loans
• Emerging Market Bonds
• Long-Term Bonds (30 year bonds)
• Closed-End Funds (CEFs) that own bonds– which tend to be more volatile given the lower liquidity and because a lot will use some leverage.

For example, a 5-year Treasury bond won’t work very well because the price swings are +/- 2% which isn’t much of a trend to ride.  The 30-year Treasury however can see 20% to 30% swings.  All of the types of bonds mentioned above are much more volatile than your typical investment grade bonds so I would not take sizeable investment positions in them without overlaying some sort of risk management system.

A trend system should also test well for various assets across multiple time frames.  At a minimum, a system should be able to beat just owning the asset as its benchmark – meaning investing in the asset and holding it for the entire period, otherwise, why trade in and out?  The system I developed and will now be implementing successfully outperformed each asset since inception, meaning it adds value as a timing system.  Here’s a chart of the SPDR Barclay’s High Yield Bond ETF (JNK) going back to its inception date of December, 2007.  The chart is coded to color green when the system says you should own JNK and red when you should be out.  This chart does not adjust for dividend distributions so it essentially shows the alpha generated by the strategy – the extra price gains in addition to the income received while you own JNK.

SPDR Barclay’s High Yield Bond ETF (JNK) – weekly chart (click to enlarge)

You can see that the system would have successfully avoided the biggest draw downs in 2008, 2011, 2014 and 2015.  You’ll also notice that the system says to be in for only about half the time.  For the other half, since high yield bonds usually get hit during periods of risk aversion across the markets (when spreads widen), I found that you can improve returns further by moving the money to intermediate-term Treasury notes (like IEI) and sitting there until the system signals it’s OK to buy back in to JNK.

Another nice aspect of using a system like this with funds is that you can take positions of larger size.  For example, I would never put 20% of a retirement oriented portfolio in 1 stock.  The volatility alone would make people uncomfortable and it’s a lot of concentrated risk.  However, I would have no problem putting 20% of a portfolio in a diversified fund.  Additionally, most brokerage companies these days have a list of transaction-free ETF’s as long as you hold them for 30 days.  I’ll be utilizing a handful of these to help reduce trading costs each time we flip in and out.

Lastly, I think the greatest benefit of utilizing a system like this is that it opens new doors for asset classes that I otherwise would have considered inappropriate investments right now (i.e. junk bonds).  I cannot justify owning high yield bonds as an investment at this stage in the cycle given the risk/reward profile (we actually haven’t owned them since the beginning of 2015).  But I have no problem buying them for a short period of time if they’re trending higher because I can rely on the system for an exit without having to make that tough judgement call of “is this just a small dip that will bounce back, or is this the start of a bigger drop…?”

It has been a very poor investment environment (long-term holding) but a great trading environment (short-term swings) for the past year and a half, and I don’t see that changing anytime soon so it’s important to adapt how I go about managing portfolios.  I have no idea what bond returns will look like over the next few years but I can say with confidence they won’t do as well as the last 5 years, especially if the bond bull market ended.  In which case, I’m confident I should be able to improve returns by incorporating a trend following system.

Thanks for following!

-Nick