If you have been a student of investing for any length of time, you have—no doubt—heard mention of the legendary efficient frontier and fabled “optimal” portfolio allocation. This raises the question—a very legitimate one: Is the efficient frontier applicable to dividend-growth investing and should we be seeking it with our portfolios?
Obviously, portfolio management—meaning the management of allocations to maximize expected return and minimize risk—is (or should be) important to all investors… including those of us engaged in DGI. However, there are lots of approaches one can take when it comes to allocations.
Which strategies work and which ones don’t? Furthermore, which ones are best suited for dividend-growth investing?
In this article, we will dive into the nuts and bolts of the efficient frontier and related topics to discern how it works and whether or not it’s a good fit for dividend-growth investing.
Specifically, we will explore:
In the end, I find that, while providing some minimal-degree of inherent value as points-of-reference, these concepts should not guide our portfolio-management decision-making process as dividend-growth investors. In fact, allowing them to do so can have materially-detrimental impacts on our long-run performance.
In other words, don’t go chasing the efficient frontier!
As with any type of analysis or predictive modeling, it is critical to understand what it is predicated on how it works. As I’ve consistently stated (see my article Dividend-Growth Investing Is All About the Process), investing success is all about the process and it’s critical that you fully understand yours. Blindly following unknown or misunderstood processes is a sure-fire recipe for investing disaster!
Let’s take a look at the efficient frontier…
The efficient frontier or minimum-variance optimization (MVO) is an approach to portfolio management that falls under the broader umbrella of modern portfolio theory (MPT).
It is all about portfolio allocation—not what is or isn’t in your portfolio.
It is an approach to portfolio management that seeks to optimize portfolio allocation based on maximizing the aggregate risk-adjusted return. In other words, it attempts to find the best allocation of holdings—the one the delivers the best return with the least risk.
Traditional portfolio allocation strategies seek to minimize the complexities involved with portfolio management by generalizing the risks associated with broad classes of assets.
For example, equities are inherently riskier than bonds. As such, more risk averse investors should skew their portfolio allocations towards bonds.
This is where we get the well-known traditional allocation guides, such as the 60/40 equity-to-bond ration.
However, these allocation models simply represent general guides, which were designed to provide general parameters for retail investors with little knowledge or understanding of investing. It’s like the lanes painted on a street—they are there to help you keep your vehicle “between the lines” and avoid driving into a ditch.
The reality is that every portfolio is unique, and these guides fail to provide effective, efficient, and actionable guidance that is targeted-to or tailor-fit for a specific portfolio (unique group of holdings). In other words, they provide a very wide lane to drive in!
Modern portfolio theory introduces the concept of minimum-variance optimization (MVO), which goes beyond the broad, cookie-cutter traditional allocation guides and attempts to provide optimized allocations for any portfolio—regardless of the specific assets that are contained within it.
How does it do this?
MVO seeks to identify the optimal portfolio allocation based on two variables: expected return and variance (aka volatility).
In essence, it identifies the lowest-variance (risk) allocation for a given level of expected return.
When you plot these points on a graph, you produce the fabled minimum-variance portfolio frontier:
This curve will always resemble a bullet shape. The point of the bullet is known as the Global Minimum Variance Portfolio (GMVP). This represents the portfolio allocation that will produce the absolute minimum amount of risk (variance) possible for the given holdings.
The portion of the curve below this point is referred to as the Inefficient Frontier and the portion of the curve above this point is the Efficient Frontier.
For any given level of risk contained within the given frontier curve, you will find two points on the curve (except at the GMVP—which is a minimum or singularity).
One of these points will correspond to an expected return that is less than the global minimum variance portfolio and the other will correspond to a higher expected return.
Obviously, if you are going to accept a certain level of risk, you would always want the allocation that provides you with the greater expected return!
Hence, the lower section of the curve would be “inefficient” and the higher section would be “efficient.”
The efficient frontier provides us with a set of portfolio allocations that produce the highest expected return for a given level of risk.
The question now becomes… which allocation should an investor select?
One method for identifying the Optimal Portfolio is to use the Sharpe Ratio.
This ratio allows us to compare different allocations to determine how much additional return we are getting for an added unit of risk. The greater the Sharpe ratio is for a given allocation, the better its risk-adjusted performance.
The point along the efficient frontier that represents the maximum Sharpe ratio is referred to as the optimal portfolio allocation.
While this all sounds good in theory, there are problems (or weaknesses) with utilizing MVO and the efficient frontier.
First, this process is entirely rearward looking. It is predicated on historical data and, as we all should know by now, past performance is not indicative of future performance. It is akin to trying to drive your car by only looking in the rearview mirror—not recommended!
Now, that is not to say that historical data is of no value or merit. A trailing PE ratio provides an investor with some degree of value—but it is limited. It must also be considered against future expectations (analysis)—meaning the forward PE and PEG.
In essence, MVO ignores future expectations predicated on sound fundamental analysis—which is never a good idea.
Second, this process is predicated on historical variance. However, not all variance is created equal and, more importantly, higher levels of variance are not always a bad thing. In fact, it can be a very good thing—especially for value investing (which is an integral part of dividend-growth investing).
A stock that has been stagnant in price over a period of a year will have a very low variance. However, this is not necessarily a positive because (a) it would have provided no positive return and (b) it would have presented no value opportunities to buy it.
To the contrary, a dividend-growth investor with a value approach wants to buy quality companies when the market has overreacted to the downside. However, this will result in a significantly-elevated variance (risk). This means an MVO approach will necessarily reduce exposure in a portfolio to this stock at the very point in time when an investor should be increasing its weighting.
[Of course, provided that this is a quality company and a forward-looking fundamental analysis indicates a real value opportunity and not a value trap!]
Furthermore, not only will an MVO approach assign a higher risk to this stock, but it will completely ignore the offsetting impact of an increasing margin of safety.
The reality is that the stock may actually be far less risky now (despite the higher variance) than it was before (with the lower variance) because of this increased margin of safety!
This is why Warren Buffett welcomes short-run variance—it provides the value investor with opportunity! This is what Buffett and Benjamin Graham refer to as the “proper market mindset!”
To learn more about the required mindset for dividend-growth success, I encourage to read my articles 3 Essential Investing Principles for Dividend-Growth Success and Paradigm Shift: The Dividend vs. Growth Investing Mindset.
Finally, a corollary to the two aforementioned weaknesses is that a minimum-variance optimization approach completely ignores the fundamentals—especially long-run ones.
Risk Parity (RP) is an investing concept that has continued to surge in interest since the financial collapse and great recession. It shares some commonalities with minimum-variance optimization but is distinct as well.
Like minimum-variance optimization, risk parity falls under the broader umbrella of modern portfolio theory (MPT). However, rather than seeking to identify a portfolio allocation that produces the greatest expected return at the lowest level of aggregate risk, RP seeks only to balance actual risk within a portfolio.
The origins of risk parity can be found in the 1950s and 1960s; however, it was Ray Dalio and Bridgewater that really pioneered the strategy in the 1990s—launching the first RP fund (their All Weather fund) in 1996.
As we noted earlier, traditional allocation guides provide very wide driving lanes. However, Ray Dalio posits that they also frequently underestimate the level of real risk in a given portfolio.
Thus, retail investors following these traditional portfolio allocations can believe they are managing their risk appropriately when, in fact, their actual portfolio risk is much greater than they thought—or would be prepared to stomach. They can unknowingly be exposed to volatility that exceeds both their personal risk profile and risk tolerance.
Dalio developed his risk-parity approach to address—and proponents of RP would argue solve—this problem.
The actual term “risk parity” was not coined until 2005 by Edward Qian in a white paper—after which it was quickly embraced by many in the industry.
Whereas MVO is predicated on two variables (expected return and variance), risk parity focuses exclusively on one: variance (risk).
Simply put, RP seeks to equalize risk across a portfolio—regardless of expected return.
Because it will inherently produce a lower expected return than those along the efficient frontier identified by minimum-variance optimization, investors implementing this strategy must often employ leverage (margin) to close the gap or even exceed it.
The theory is that if the RP portfolio can be leveraged to produce a higher expected return (factoring in the cost of leverage) than the MVO approach for a given level of risk, then it is more efficient and provides a greater risk-adjusted return.
In addition to all the problems inherent in the MVO approach, the risk parity approach adds a number of additional issues and concerns.
First, risk-parity requires leverage—not optimal for retail investors. Because you are overweighting less-risky assets, you are reducing the portfolio’s aggregate expected return. Overcoming this inherent weakness requires leverage.
Second, the RP portfolio must have a higher Sharpe ratio than other allocations (e.g., the global minimum variance allocation, optimal portfolio allocation, or traditional allocation). If not, then the risk-parity portfolio will underperform them regardless of how much leverage is applied.
Third, asset classes (however defined) do not provide the same risk-adjusted returns.
Fourth, asset classes are arbitrary classifications and, therefore, every portfolio can be defined as risk-equalized by simply redefining those classes.
Fifth, risk-parity portfolios are highly-vulnerable to changes in asset class correlations, such as equity-bond correlations. For example, with an RP portfolio, total risk can more then double when going from a negative to positive correlation (see The Risk in Risk-Parity Strategies).
Sixth, any potential benefit inherent in a risk-parity strategy has been greatly diminished as retail investors have improved their level of portfolio diversification over the past two decades—thanks to the availability of greater information, tools, and investing options.
Finally, if you are using a risk-parity investment product (e.g., a fund) rather than implementing the strategy on your own, you are likely to encounter material expense fees.
In order to demonstrate what an efficient frontier actually looks like, as well as different approaches to portfolio allocations, let’s look at a real-world example.
We’ll use our site’s PIP (passive-income portfolio). Since our portfolio includes 90+ positions, let’s simplify the process and focus on our financials sector and, more specifically, our real estate sub-component—consisting of 9 well-diversified (across the industry) holdings: $RLJ, $LMRKN, $APTS, $BRX, $MPW, $UMH, $LTC, $LAND, AND $CUBE.
To perform this type of analysis, you will need to use the Solver function in Excel. There are a number of great (free) resources out there to learn how to do this (e.g., YouTube), so I won’t address the process in this article… just the results.
Here is what it the data looks like for our holdings:
Once we crunch all the data, here’s what the Efficient Frontier, Global Minimum Variance Portfolio (GMVP), Optimal Portfolio, Risk-Parity Portfolio, and Actual Portfolio look like when graphed:
Now, let’s look at each of these 4 allocations in detail:
At first glance, many of you may be wondering why in the world our current allocation is what it is! I mean it looks like we are taking-on more risk for less return… right?
Well, not so fast!
Our approach to portfolio allocation management is predicated on an entirely different process than that of minimum-variance optimization or risk parity.
Here are some critical distinctions:
We only include quality companies we are comfortable with owning forever and accepting their risk. As such, we are not handcuffed by an approach that is constrained by historical variance.
We have a very long investing time-horizon (in most cases, forever).
We employ a forward-focused fundamental approach. While we obviously consider historical data, we discount it in terms of how it influences our decision-making process moving forward.
We are, at our core, value investors with regards to this portfolio. Yes, it features a dividend-growth strategy; however, we are looking for dividend-growth at a value—with a margin of safety based on intrinsic value.
This portfolio is in the building (or accumulation) phase.
Based on the all the above distinctions, we seek to add to positions based on opportunities to increase our yield-on-cost (YOC) and lower our cost basis—taking advantage of short-run market overreactions and our time-arbitrage edge (with an eye on ever-changing long-run company fundamentals). We rank our opportunities each week based on formula that highlights the biggest/best opportunities.
This enables us to fully leverage cost-averaging as we build out the portfolio—deploying our capital in the most efficient manner.
Furthermore, over a long period of time, this process will produce an end-result (portfolio allocation) that will steadily move towards a higher risk-adjusted aggregate return.
If we chased the efficient frontier, we would be forcing allocations that don’t fit the current market environment. In other words, we would be adding where it doesn’t make sense—deploying our capital less efficiently.
For example, if we were to chase the optimal allocation, we would need to add to our positions in $LMRKN (significantly) and $CUBE, while holding-off on adding to our position in $RLJ. However, there is currently far greater opportunity in $RLJ than $LMRKN or $CUBE.
This doesn’t mean we don’t want to continue building our position in $LMRKN and $CUBE—we most certainly do. But it does mean that we want to do it when it makes sense!
Remember, MPT strategies will underweight holdings when volatility increases (such as a short-run market overreaction to the downside). However, this is precisely when we want to step in and buy—when everyone else is running for the exits and we see value!
Modern Portfolio Theory—whether from a minimum-variance or risk-parity perspective—is a trendy topic. While these approaches provide advantages over traditional allocation approaches, they are not without significant weaknesses.
These strategies are best-suited for shorter-term, growth investing—not long-run dividend-growth investing. The longer your investing time-horizon is, the less value these portfolio management approaches will provide you with.
At the end of the day, investing—regardless of style or approach—is all about your process. It is critical that you understand your process and stick with it, as well as understanding the process behind other options you may be considering.
MPT simply doesn’t match our preferred process, nor do we feel it fits particularly well with any dividend-growth investing strategy.
This doesn’t mean these strategies for portfolio allocation are inherently bad. Rather, it simply means they are not a good fit for our process. We’ll stick with our process and avoid chasing fads—no matter how popular they may be at the moment.
I encourage you to do your homework and understand these MPT strategies before you start chasing the efficient frontier!
Specifically, it is important to understand their inherent flaws, including these three critical weaknesses:
I am reminded of a quote from Seth Klarman regarding Buffett’s purchase of The Washington Post:
“It is helpful to note that Buffett did not consider whether The Washington Post was a component of a stock market index, or about to be added to one. He did not weigh the market capitalization of the company or its daily trading volume in his purchase decision… He most certainly did not evaluate the stock’s beta or use the capital-asset pricing model or consider whether its purchase would move his portfolio to the efficient frontier. He simply valued the business and bought a piece of it at a sizable discount.”
Dividend-growth investing is not rocket science. While Modern Portfolio Theory, with its pursuit of optimization, provides copious amounts of opportunity for entertaining mental gymnastics and academic exercises, investing doesn’t need to be that complicated!
Keep it as simple as possible. Doing so will increase the likelihood that you will (a) stick with it and (b) achieve long-run success!
Always remember, investing involves substantial risk of loss and is not suitable for everyone. The valuation of investments may fluctuate, and, as a result, you may lose substantial amounts of money. No one should make any investment decision without first consulting his or her own financial adviser and conducting his or her own research and due diligence.
You should not treat any opinion expressed on the Wicked Capital website as a specific inducement to make a particular investment or follow a particular strategy, but only as an expression of opinion for entertainment purposes.
The opinions are based upon information we consider reliable, but neither Wicked Capital nor its affiliates, partners and/or subsidiaries warrant its completeness or accuracy, and it should not be relied upon as such.
Past performance is not indicative of future results. Wicked Capital does not guarantee any specific outcome or profit. You should be aware of the real risk of loss in following any strategy or investment discussed on this website.
As noted above, strategies or investments discussed may fluctuate in price or value. Investors may get back less than invested.
Investments or strategies mentioned on this website may not be suitable for you. The material presented does not take into account your particular investment objectives, risk tolerance, financial situation, or needs and is not intended as recommendations appropriate for you. You must always make an independent decision regarding investments or strategies mentioned on this website. Before acting on information provided on this website, you should consider whether it is suitable for your particular circumstances and strongly consider seeking advice from your own licensed financial or investment adviser.