Mad Money Host Goes Diversification Hunting
If you have ever watched CNBC's Mad Money, you know that host Jim Cramer leaves a memorable impression. A typical episode involves so much set activity, visuals and sound effects that Jim is usually sweating half way through the show. My personal favorite is Jim's sound board, which he uses early and often to respond to caller's questions throughout the broadcast.
In a segment entitled "Am I Diversified?" the caller rattles off five stocks in his or her portfolio and Jim makes the pronouncement, in real time, about whether he believes the caller's portfolio is "diversified." This mini-case study is all about portfolio diversification, including what it is, whether you can achieve it on the equity (stock) portion of your portfolio with just five stocks, and why you should even care if your portfolio is properly diversified.
First, a quick definition of portfolio diversification courtesy of Investopedia, a website dedicated to investor education:
A risk management technique that mixes a wide variety of investments within a portfolio. The rationale behind this technique contends that a portfolio of different kinds of investments will, on average, yield higher returns and pose a lower risk than any individual investment found within the portfolio.
At first glance, you might think more is better, but what really counts is whether your stocks are "different" from each other as measured by correlation. We'll cover correlation later in this blog post, but now we need to create test scenarios (portfolios) to see if holding just five stocks for the equity portion of a stock and bond portfolio achieves true diversification.
Selecting the Test Portfolios
In selecting the test portfolios I first needed a benchmark portfolio. I used the classic 60/40 stocks to bonds balanced portfolio, but kept the bond portion the same across all test cases in order to study diversification on just the stock portion of each test portfolio. The Vanguard S&P 500 Index Fund [VFINX] was selected as a proxy for the U.S. stock market and the Vanguard Total Bond Index Fund [VBMFX] as a proxy for the U.S. bond market.
Next, for Jim Cramer's test balanced portfolio, I chose five large cap stocks, each from different S&P 500 industry sectors. There are 10 broad sectors within the S&P 500 index: financials, basic materials, technology, industrials, consumer discretionary, energy, telecommunications, healthcare, utilities and consumer staples. While I don't know Mr. Cramer's exact diversification criteria, I am fairly confident that he wants to see individual stocks from different S&P 500 sectors as a first-level diversification screen to avoid any sector overlap.
The five Jim Cramer stocks I chose for this hypothetical test include (1) JPMorgan Chase (NYSE:JPM) (Financial), (2) IBM (NYSE:IBM) (Technology), (3) Chevron (NYSE:CVX) (Energy), (4) Johnson & Johnson (NYSE:JNJ) (Healthcare) and (5) Southern Co. (NYSE:SO) (Utilities). These five stocks meet the different S&P 500 sector criteria and are generally considered blue chip companies, which means they are stable, well-run enterprises with proven business models.
Given that there are ten S&P 500 sectors, why those five companies and sectors? The short answer is beta. A stock's beta is the measure of how much that stock's price movement results from the overall movement of the stock market in general. The measure is scaled against 1 or 100% for the entire stock market, or the Vanguard S&P 500 Index Fund [VFINX] proxy in our test case. Using sector Exchanged-Traded Funds (ETFs), I obtained the 10-year average beta for each of the S&P 500 sector ETFs using Quantext Portfolio Planner. The results are provided in Diagram 1.
Next, I selected two sector representative companies that had betas higher than the market (100%), indicating their prices are more volatile than the market. These sectors included financials and technology. So if the market rises (or falls) 1% in a given day, on average the financial and technology sector ETFs will rise (or fall) more than 1% given their respective betas of 134% and 122%. I then selected the sector ETF with the 10-year average beta closest to that of the market (100%) which turned out to be the energy sector. And finally, I selected two sectors (healthcare and utilities) with less volatile betas (i.e. under 100%) to round out the experiment. The resulting Jim Cramer balanced test portfolio is provided below in Diagram 2.
The idea is to select five stocks that, when combined together have roughly (and I mean roughly), similar long-term beta characteristics to the S&P 500 index, or in our test case the Vanguard S&P 500 Index Fund [VFINX]. Now we have our Benchmark Balanced portfolio and our Cramer Balanced test portfolio to compare portfolio diversification behaviors.
I also created third test portfolio, the Alternative (ALT) Balanced portfolio to include all 10 representative sector companies in smaller allocation percentages (4% versus 12%). I did this because many advisors limit their single stock positions to no more than 4% of the total portfolio to avoid concentrated stock risk. This is known as unsystematic (idiosyncratic) risk, or the factors other than general stock market price action that affect a specific company's stock price. I apologize for the use of all the financial jargon, but you'll see why shortly I had to define these terms.
Anyway, the resulting third test balanced portfolio includes our fixed 40% allocation to bonds, another 40% to large cap blue chip companies in increments of no more that 4%, leaving 20% to allocate to other asset classes. For fun, I allocated 8% to a small cap index fund to see if this asset class adds any diversification benefit. I also allocated the remaining 12% to alternative asset classes that in general have a long-term history of being solid stock/bond diversifiers. The resulting third test portfolio is provided below in Diagram 3.
|Click to enlarge|
|Diagram 3: Third test portfolio (Alt Balanced) with 10 companies in 4% percentage allocations|
We have our three test portfolios and are now ready to analyze their diversification characteristics.
Reporting the Results
Given my stated blog mantra to provide "more visual beef and less jargon-laden dribble," here's where we are going to use multimedia for the show-and-tell of the test results. I've tailored my portfolio evaluation and retirement income analysis template to accommodate our case study and report on our three test portfolios:
- Benchmark Balanced portfolio
- Cramer Balanced portfolio
- Alt Balanced portfolio
A typical Koch Capital portfolio evaluation and retirement income analysis package includes (1) a brief video summary, (2) an online results presentation, (3) a detailed report package available for further analysis or reference if needed (I call this my Mobile Planner) and (4) optional links to additional topical resources. So here are the links to the web-based report package elements.
I recommend starting with the Jim Cramer Am I Diversified summary video below.
Assuming you hung in this far with the case study and watched the video summary, I want to reiterate several key findings from the analysis.
First, while all three balanced portfolios achieved positive average historical returns from 11/30/2001 through 11/30/2011, a very tough market decade by any measure, the forward return and standard deviation (volatility) estimates in blue below suggest that the test portfolios holding concentrated single stock positions will continue to out-perform the Benchmark Balanced portfolio on a risk-adjusted basis. This projected result, in my opinion, comes from better diversification as demonstrated by the improving Hist. Beta (decreasing) and Hist. D.M. (increasing) scores.
On the historical beta score, our Benchmark Balanced portfolio has a 60% allocation to the S&P 500 stock index (since the Benchmark Balanced portfolio as well as the other two portfolios hold a fixed 40% allocation to bonds) and a beta score of 60%. This means it's less volatile than the S&P 500 all-stock portfolio, which has a beta of 100%. So the beta scores of less than 60% for the Cramer Balanced and Alt Balanced indicate these portfolios are even less sensitive to price fluctuations in the overall stock market than our Benchmark Balanced portfolio.
The other diversification score is D.M. (Diversification Metric courtesy of Quantext Portfolio Planner) or non-market beta, as I often refer to it. It measures the "differences" or correlations between the company-specific portion of the returns of an individual stock (or mutual fund) regardless of what the overall stock market is doing. From my own experience, I tend to rely heavily on the historical D.M score since I find an investor can construct better long-term, stable return portfolios by obtaining as high as possible D.M. score while limiting beta to an acceptable client risk tolerance level.
So why does all this diversification stuff matter? In a nutshell, I believe it matters because it provides the investor with better odds of constructing a stable return portfolio that has the highest probability of providing a sustainable cash flow in retirement. The hypothetical retirement distribution glide paths for the three test portfolios are provided below for reference.
|Click to enlarge|
|Diagram 5: Hypothetical glide paths assuming retirement at 65 and flat $150,000 withdrawals|
Again, let's assume for our hypothetical example we seed each portfolio with one million dollars in 2011, let each grow for nine years until 2020 when Mr. Cramer turns 65, then withdraw a flat $150,000 annually to deliberately run down each portfolio in a relatively short retirement period.
We see the widening spreads between the 90% chance of making it to a certain age (green) and the 50% chance of not running our money by a certain age (yellow) as we move through the three test portfolios with improving diversification scores. Moreover, Jim Cramer improves the "certainty" that the Cramer Balanced and Alt Balanced portfolios will not run out of money (90% probability) by ages 78 and 80, respectively, which are significantly better odds than our test Benchmark Balanced portfolio's 10% chance of depletion by age 75.
No fiduciary-bound advisor would recommend a portfolio consisting of 5, or possibly 10, individual stocks even if it makes sense in theory. There's simply too much severe event risk to the client if just one company that comprises 12% of your client's portfolio blows up, even if the likelihood of such an event is rare. In addition, there's career risk for the advisor who makes the bad recommendation. Hedge fund and mutual fund managers have more lee way to build concentrated portfolios, but limit who can invest and generally have to disclose up front the risks involved in holding concentrated positions.
As a general practice example, client portfolios of less than $200,000 managed directly by Koch Capital would typically consist of primarily mutual funds and ETFs, while client portfolios of over $1,000,000 may consist of primarily individual positions. Portfolios in between those approximate dollar limits may consist of all of the above depending on the client's needs.
So why does it appear, at least theoretically, that holding a limited number of individual stocks improves diversification versus holding a basket of 500 stocks? Again, in a nutshell, it's all about correlation. Even though correlation values can be as volatile as stock prices, we can see from the diagram below that, in general, individual stocks tend to offer better portfolio diversification benefits via lower correlations between two positions (see pale yellow VFINX and VEXMX values versus light blue JPM, IBM, CVX, JNJ and SO values), while broad baskets of stocks (index mutual funds and ETFs) tend to dull the correlation effect (a.k.a. higher correlations, less diversification) in my opinion.
|Click to enlarge|
|Diagram 6: Correlation data courtesy of Koch Capital and Quantext Portfolio Planner|
This is why portfolio mavens like David Swensen, who manages the Yale endowment, and Warren Buffett of Berkshire Hathaway fame, can achieve such stable return portfolios. They each select individual companies as well as asset classes that have low or no long-term correlations with the other positions within their respective portfolios. It's very telling that these two investors can achieve volatility ratios (standard deviation divided by return) at 1.0 or lower (see Volatility row gauge in Diagram 4 above) using very little bond and cash allocations. Fortunately for all of us above-average advisors, as well as the intrepid independent investor, more and more robust, affordable analytical tools are becoming available over the web so portfolio builders can intelligently mitigate some of the "risk" in "idiosyncratic risk" to achieve the stable return benefits afforded by holding individual stock positions.
Finally, while I disagree with Jim Cramer's primary focus on active stock trading, he deserves a big boo-yah for pointing out the importance of diversification to his Mad Money viewers. More over, thank you Mr. Cramer for getting the young investor interested in participating in the stock market again.
Additional disclosure: Performance data quoted represents past performance results for a hypothetical portfolio during the corresponding time period. Past performance is not indicative of future results. Actual returns depend on an investor’s situation and may differ from those shown. No assumption should be made that any investment strategy or security referenced herein will be profitable or that will equal the indicated performance. Nothing presented herein is or is intended to constitute investment advice, and no investment decision should be made based on any information provided herein. Under no circumstances does the information contained within represent a recommendation to buy or sell any particular security or pursue any investment strategy. Asset allocation and portfolio diversification cannot assure or guarantee better performance and cannot eliminate the risk of investment losses. As with any investment strategy, there is potential for profit as well as the possibility of loss. Additional disclosure available here.