The “Three Fund” Portfolio Part 1 – Domestic Stocks


Recently, I reviewed model “risk-based portfolios” (i.e. conservative, moderate, aggressive) from First Trust and Blackrock. These companies promote their portfolios to investors and financial advisors under the guise of “modern portfolio theory” diversification. Typically, they recommend between 10-16 in-house ETFs. These investment shops update their portfolios quarterly to overweight the hottest sectors or factors (e.g. growth over value). These models include allocations to domestic stocks, foreign stocks and bonds.

Surprisingly, I found that NONE of the models beat a simple benchmark combination of VTI, VEU and AGG on ANY timeframe. VTI is the Vanguard Total US Stock Market ETF. VEU is the All-World ex-US Stock ETF. And AGG is the Total US Bond Market ETF (which I slightly prefer over Vanguard’s BND – see Part 3 of this series).

Immediately, I decided to see whether I could build a better model than a benchmark allocation to VTI, VEU and AGG. I used Portfolio Visualizer’s Backtest Portfolio function and focused on the VTI component.

In Favor of Dividend Growth

I had noticed that VIG, Vanguard’s Large Cap Dividend Growth ETF, seemed to offer similar returns as VTI with less volatility. I like the idea of investing in companies that are profitable enough to have a history of increasing their dividends. These tend to be quality stocks where management is confident about the future. However, vast sector differences exist between VTI (heavier on Tech) and VIG (heavier on Industrials).

Index investing works due in part because momentum works. Sectors that are performing well grow larger as a percentage of the index. So when your portfolio has major differences in sector composition to the index, you are essentially betting against momentum. In other words, you take the position that your overweight sectors will outperform the index and your underweight sectors will underperform the index.

I devised a solution to this problem which I will discuss below. However, while the S&P 500 Large Cap companies comprise most (~85%) of the stock market, they don’t encompass the total market. According to Fama’s original research, smaller companies outperform lager companies over time. Thus, using VIG and omitting smaller companies seemed imprudent.

So instead of VIG, I chose DGRO, the iShares Core Dividend Growth ETF. DGRO selects quality dividend growers from the Large, Mid and Small Cap universes. Specifically, it chooses from the Morningstar US Market Index composed of nearly 2000 of the most liquid US stocks. Therefore, DGRO serves as a better “Total Stock Market” replacement than VIG. To wit, VIG holds 183 stocks while DGRO holds 477.


A comparison of DGRO (“Portfolio 1”) to VTI since July 2014 follows below. Notice the slightly better return at a slightly lower standard deviation:

It’s a fairly simple matter to equalize the sector weights between DGRO and VTI. I have created an easy Google Spreadsheet comparing the weights of DGRO and ITOT. ITOT is a total market ETF virtually identical to VTI. Once a year, I simply manually input the weights and the spreadsheet will calculate how much of DGRO to buy (currently 74%). In addition, the spreadsheet calculates how much of the Vanguard total market sector ETFs to buy (currently 11% VGT, 5% VCR, 6% VOX and 4% VNQ).

I ran the new portfolio in Portfolio Visualizer. Note that the numbers won’t be accurate “point in time.” Sector weight differences between DGRO and ITOT would have changed over time since 2014. DGRO is reconstituted every December and rebalanced quarterly, and thus experiences about 25% annual turnover affecting sector weights. But for illustration, you can see the potential improvement. Annual return, Sharpe and Sortino have all increased, while Standard Deviation barely budged:


DGRO, either alone or supplemented with sector ETFs to equalize sector exposure to VTI, is a better core US stock holding in a risk-based model than VTI alone.

Next up, I will look at foreign stocks in Part 2.

Note: With sector equalization, my “three fund” portfolio is actually eight funds. The three main funds account for approximately 80% of it. Further, I personally use a second sector equalization step in building my portfolio. For the password to my personal Aggressive Model, please contact me.

0 0 vote
Article Rating
Notify of
Inline Feedbacks
View all comments