Build Your Own Investment Portfolio, Part 8

Teresa, March 23, 2008 @ 7:16PM ET | Link | RSS | Read via Email | 28 Comments  

In Part 7, we reviewed two principles upon which portfolios are built.

Let’s construct a portfolio for our archetype client, defined in Part 5: professional, owns a house, 15-20 years from retirement, takes full advantage of an employer-matched plan, has 3 dollars in the retirement account for every dollar in the trading account.

The Core Portfolio

Employer-matched plans tend to offer investment choices among traditional asset classes. Virtually all plans offer the equivalent of the following four exchange traded funds:

  1. U.S. Equities: S&P 500 (SPY), inception date 1993.01.29
  2. International Equities: MSCI EAFE Index Fund (EFA), inception date 2001.08.14
  3. U.S. Treasury Bonds: Lehman 7-10 Year Treasury Bond Fund (IEF), inception date 2002.07.22
  4. U.S. Treasury Inflation Protected Securities: Lehman TIPS Bond Fund (TIP), inception date 2003.12.04

Plans often limit rebalancing frequency; therefore, the Core Portfolio is rebalanced once per month, on the first trading day of each month.

Going with our “no rear view mirror” policy, the portfolio formation date was April 1, 2004 when TIP, the last of the exchange funds required, had been trading for one full quarter.

23-coretotalreturn.gif
Total Return (%): April 1, 2004 - March 20, 2008

The total return over four years was 34%, or each dollar at portfolio formation would now be $1.34 to produce an “average” annual return of 8.5%. While these numbers do not set the world on fire, we achieve the twin goals of 1. taking advantage of employer-matched funds, and 2. maximizing the effects of compounding in a tax-deferred account, doing it in an orderly manner without putting our retirement funds in mortal danger. Easy does it.

23-corevolatility.gif
Monthly Returns (%): April 1, 2004 - March 20, 2008

The month-to-month changes in account value are consistent without big months to the downside, making it easy to stay the course and avoid drastic measures generally taken in the heat of market madness. (See volatility of returns.) When it comes to long-term investing, slow and steady wins the race. Mission accomplished.

The Satellite Portfolio

Our archetype client has one dollar in his trading account for every three in his retirement account, along with some “mad money”. The client has a full life, and as such, may not have time (or even want to) pick stocks for trading.

We harnessed the power of diversification and the efficacy of our rebalancing algorithm to construct a high-octane version of the Core Portfolio to take advantage of alternative asset classes and 1:1 leverage (50% margin) with the following exchange traded funds:

  1. U.S. Equities: Vanguard Extended Market ETF covers small- and mid-cap stocks (VXF), inception date 2001.12.27
  2. International Equities: MSCI Emerging Markets Index Fund (EEM), inception date 2003.04.07
  3. U.S. Treasury Bonds: Lehman 20+ Year Treasury Bond Fund (TLT), inception date 2002.07.22
  4. Currencies: CurrencyShares Swiss Franc Trust (FXF), inception date 2006.06.21
  5. Commodities: S&P GSCI(TM) Commodity Indexed Trust (GSG), inception date 2006.07.10

Trading accounts can rebalance as often as necessary, yet it is important to contain transaction costs. To achieve economy, we selected a single ETF that best embodies the characteristics of each asset class and rebalanced the Satellite Portfolio on the first trading day of each week.

Going with our “no rear view mirror” policy, the portfolio formation date was October 23, 2006 when GSG, the last of the exchange funds required, had been trading for one full quarter.

23-sattotalreturn.gif
Total Return (%): October 23, 2006 - March 20, 2008

The total return over 17 months was 47%, or each dollar at portfolio formation would now be $1.47 to produce an “average” annual return of 33%. While these numbers indeed set world on fire, we must emphasize the fact that the portfolio is actually quite balanced, sailing through recent market turmoil without causing sleepless nights.

How do we measure up?

In Part 3, readers were asked which of the famous funds they would buy. Most probably selected the CGM Focus Fund which “may borrow from banks in an amount not to exceed one-third of the value of its total assets and may borrow for temporary purposes from entities other than banks in an amount not to exceed 5% of the value of its total assets.”

23-satvscgm.gif
InVivo Satellite (% total return) vs. CGMFX (% return, ex-dividend)

We did not calculate the total return for CGMFX because it was not necessary for the purpose of this exercise. Readers can plainly see that there are basically two ways to get to the goal: the Tortoise or the Hare.

We know from the world of behavioral finance that making decisions under risk is not one of the strengths of human nature. While some CGMFX investors would have been able to stick to it through thick and thin, the resolve of most unit holders would have been severely tested in recent months.

23-cgmfxvolat.gif
Weekly Returns (%, ex-div): October 23, 2006 - March 20, 2008

23-satvolat.gif
Weekly Returns (%): October 23, 2006 - March 20, 2008

Even though our Satellite Portfolio used more leverage than CGMFX (50% vs. up to 33.3%), the volatility of returns was dramatically lowered by the synergistic effect of portfolio diversification and the use of our proprietary allocation algorithm.

The bottom line is this: There is no better time than NOW to get off the investment roller coaster. You deserve some smooth sailing…

Build Your Own Investment Portfolio, Part 7

Teresa, March 20, 2008 @ 9:30AM ET | Link | RSS | Read via Email | Start a Discussion  

In Part 6, we analyzed what investors might consider to be an advanced article on portfolio construction and asset allocation.

Let’s review the two core principles upon which we build a portfolio:

Principle #1: Asset Classes Defined According to Function

David Swensen said it best, stating “Careful investors define asset classes in terms of function, relating security characteristics to the role expected from a particular group of investments.” Furthermore:

Purity of asset class represents a rarely achieved ideal. Carried to an extreme, investors define dozens of asset classes, creating an unmanageable multiplicity of alternatives. While market participants disagree on the appropriate number of asset classes, fewer tend to be better. Portfolio commitments must be large enough to make a difference. Committing less than 10 percent of a fund to a particular type of investment makes little sense; the small allocation holds no potential to influence overall portfolio results. Investors trade off the benefits of precise definition of an asset class with the costs of employing large numbers of classes.

Functional attributes play the dominant role in defining asset classes, with structural and legal characteristics taking secondary positions. Asset class distinctions rest on broad sweeping differences in fundamental character; debt versus equity, private versus public, liquid versus illiquid, domestic versus foreign, inflation sensitive versus deflation sensitive. Ultimately investors attempt to group like with like, creating relatively homogenous groups of investments that provide fundamental building blocks for the portfolio construction process. — David Swensen, Pioneering Portfolio Management

Assets expected to react in similar fashion to a given economic environment belong to one asset class. An all-weather portfolio should diversify among asset classes that are expected to respond differently to an ever-changing economic environment.

Principle #2: Use a Proven Framework to Allocate the Assets

In Part 6, we explained the shortcomings of equal weight allocation and introduced classic frameworks such as CAPM, APT and MPT. The past decade has seen the introduction of advanced asset allocation frameworks (see Part 2, further reading) such a risk parity portfolios.

The InVivo tradition of excellence springs from a commitment to research and development, mainly because my own investment and trading performance is the chief beneficiary of my work. It should come as no surprise that we developed our own dynamic asset allocation program.

The skeptical reader wrote to say that he does not trust “black boxes”, as if the factors that cause hedge funds to blow up during every market debacle could even remotely be connected to an asset allocation program. The fact is that speculative statistical arbitrage of exotic debt instruments leveraged 30 times has nothing to do with an algorithm that calculates the allocation of a person’s unleveraged investment portfolio.

While we cannot provide the details of our rebalance algorithm, suffice to say that it works. I would not use it on my own money if it didn’t, but I wanted to show readers how something as mundane as deciding how much of each asset class to buy can make a big difference.

We devised an elegant (always-long, no cash) experiment to demonstrate the effectiveness of our rebalancing program. We allocated funds within one asset class by using applying our algorithm on the Select Sector SPDRs and compared the results with the performance of SPY, the S&P 500 ETF.

19-spdrexperiment.gif

Select Sector SPDRs are ETFs that divide the constituent stocks of the S&P 500 Index into nine sector index funds. Any out- or underperformance in the SPDR portfolio can be directly attributed to our algorithm. Our SPDR allocation dampened the ups and downs of the market roller coaster and outperformed SPY over the long haul.

21-spdr.gif

Furthermore, the volatility of returns (discussed in Part 3) was reduced considerably.

21-spy.gif

The lesson here is that we must focus on principles and not get caught up splitting hairs. A strong portfolio requires exposure to equities and several other asset classes in order to achieve diversification enough to deal with unanticipated inflation, deflation and currency risk. Portfolios must be rebalanced with discipline using a proven method. It is not enough to put lipstick on a pig; we must make it fly.

In the next and final installment of the series, we will put together a real-world, all-weather portfolio for our archetype client and show you the results.

CLICK HERE to read Part 8.