Wednesday, September 30, 2015

How to properly weight dividend portfolio holdings

There are different ways to weight a dividend portfolio. I am going to examine the three most popular methods in this article. Then I am going to reflect on the method I use.

The first method is weighting portfolios based on market capitalization, and then adjusting weights based on free floats. The logic is that as a company becomes more valuable, it should have a higher weight in a portfolio, whereas a company that is less prosperous should have lower weights since its capitalization is lower. This method is preferred by many index funds, as it makes it easier to just buy a basket of several hundred or thousand securities, and then passively hold them. It is viewed as a self-cleansing mechanism, where prosperous companies gain higher weightings, while less prosperous companies are eventually flushed out. The dangers behind this method is that a speculative company with no material profits and elevated valuations could get a higher weighting due to stock price being bid up by speculators. This happened with a lot of indexes such as the S&P 500 during the tech boom in the late 1990s, when they added companies like Yahoo! (YHOO) at several hundred times forward earnings. The results were disastrous, and pulled down the expected returns for all index investors. On one side, it is a good idea to give higher weighting to the companies that are prospering. On the other side however, you could end up in a dangerous situation where the most successful companies will get large and command a large piece of the pie. However, these large companies might end up dominating that portfolio. In fact, the 50 largest components of S&P 500 account for 47% of the portfolio.  Apple (AAPL) accounts for almost 4% of the portfolio. Through July of 2015, just six companies in the S&P 500 accounted for most of the index gains. I wonder if this is a repeat of 1999 - 2000, or not.

The second method utilized by some funds is to weight holdings based on dividend yields. A prime example includes the S&P Dividend Aristocrats ETF (SDY), which assigns higher portfolio weights to higher yielding components. This results in a higher yield for the portfolio, and potentially higher returns. Unfortunately, this also increases the risk that riskier companies with unsustainable high dividends have higher weights. A company that is about to cut dividends will usually spot an above average dividend yield. This would result in a higher portfolio weight. If the company then cuts dividends, investors would suffer higher than average losses.

A third method involves equally weighting each component. This method allocates an equal chance of success to each individual component, regardless of size. When applied to the S&P 500, this method ends up doing better than a market-cap weighted index by approximately 2.25% per year over the past 50 years. Each component is given equal chances of pulling the cart. However, it also requires re-balancing every so often, which could be costly in a taxable account.

So how do I weight individual stocks in my portfolio? The answer is that it depends on the situation.

For example, when I deal with a large lump-sum, I would create a portfolio from scratch, and allocate the money equally between a certain number of companies. For example, back in early 2013 I rolled over an old 401 (k) into an IRA. I put the money to work in an equal weighted portfolio of twenty dividend paying stocks.

In my regular investing however, things are not always as obvious. When I allocate capital, I allocate it in my best ideas at the time. I strive to maintain an equal allocation to every component in my portfolio. Since I have invested money every month for the past 8 years, this is not always feasible. As I build out my portfolio, those best ideas change. Therefore, I would not force an equal allocation to my portfolio holdings regardless of valuation, prospects or dividend sustainability.

I mostly put money into companies that have passed through my screen, which have attractive prospects going forward, and which have attractive valuations. As you can see, the main driver behind allocating capital is availability of quality ideas at attractive valuations. Equal weighting is something I try to achieve, but it is secondary in nature due to the changing landscape of investment opportunities. In other words, it makes no sense to add to a position that doesn't make sense anymore, just for the sake of maintaining an equal allocation.

The only reason when I use portfolio weights as a reason to avoid adding to a position is when this stock is one of the five or ten highest stock positions in my dividend portfolio.

I followed allocations and diversification as I was building out my portfolio – though not very strictly. It was not unheard of when 3 of my holdings accounted for 12.50% of portfolio, but 25% of income. However, I did not worry because I kept adding money monthly, and the relative weights of those components decreased over time. As a result, I did not add any new funds to those companies for a while. In another example, I view Johnson & Johnson as an outstanding dividend growth stock to hold for decades. However, it is one of my largest positions. If I found Johnson & Johnson to be a better value than another company in which I have a lower allocation, I would choose the company with the lower allocation. Therefore, despite the fact that I find Johnson & Johnson to be a good value today, I do not expect to add much money there. Same goes for Kinder Morgan Inc (KMI) - it is a bargain today, which I cannot add to because it is my largest position. Managing downside risk is very important for me.

Since I have a certain dollar amount to put to work each month, I tend to build a position in a stock gradually. If the company is overvalued or if something changes my assessment of the company, I don’t add to it, but will buy something else. As a result, some companies will end up small positions for me. A very good example is M&T Bank (MTB), which I initiated in 2008. The company had a fantastic earnings and dividend growth record, was well run. In addition, it was attractively valued at the time too. However, as the financial crisis hit and the bank froze dividends after receiving TARP money, I stopped adding money there. I have held on to my exposure there, but since I have been adding money for 7 years since then, this stock is just a footnote in my stock portfolio. A footnote that has returned close to a third of my purchase price in dividends alone since 2008, and has also doubled in price. This is an example of having your cake, and eating it too.

Other reasons why I would stop adding to a position include high valuation, or diminished prospects. For example, I first initiated a position in Visa (V) in 2011, when it was selling at 20 times earnings. I noticed the company when it raised dividends in 2011, and when I noticed that Warren Buffett was owning it. However, the stock was overvalued until 2014, when I added a little more at 20 times earnings.

In another example, I routinely added to shares of Procter & Gamble (PG), up until early 2013. Unfortunately, the stock has been unable to grow earnings per share since 2008, which is why dividend growth is slowing down and future dividend growth is going to be that much more difficult to achieve. As a result, I will not be adding more money to this stock.

In this article I discussed three popular ways to weight dividend portfolio holdings. In addition, I discussed my unique spin to portfolio weighting. My method is dependent on availability of good ideas on sale, amount of capital available and existing portfolio weights.

What is your process of weighting companies within your dividend portfolio?

Full Disclosure: Long V, PG, MTB, KMI, JNJ,

Relevant Articles:

Is your dividend income riskier than expected?
Dividends Offer an Instant Rebate on Your Purchase Price
Does Fixed Income Allocation Make Sense for Dividend Investors Today?
Should you follow Buffett’s latest investments?
My Dividend Portfolio Looks Much Better than I Expected


  1. I kinda like the first method here. However I just started focusing on dividend purchases so I am not really balanced at all. I am more focused on getting stocks at fair prices vs a balanced portfolio right now. "The dangers behind this method is that a speculative company with no material profits and elevated valuations could get a higher weighting due to stock price being bid up by speculators." When I pick a dividend stock, I usually try to pick one with a long history of increased payouts, and one that is ingrained in our society. So I am not sure how much of a drawback this would be, I don't feel I would be gathering speculative stocks with dividends in mind.

    Do you feel I should be focused on a balance approach so early on? If not, when do you feel it would be wise to focus more on a long term balanced approach?


  2. DGI,
    Interesting review of different ways to "weight" the differnt holdings in a portfolio. I personally approach this from the point of view of starting out with around 0.5% of the portfolio value as a "starter" position in a company that has passed all my DG screening criteria. That sounds "small", but I have a very large portfolio (5 million), so it turns out to be about 25K. If the stock remains undervalued/ promising from a DG perspective, I will slowly add to it. I have been doing that this year as stocks have corrected down, and opened postions in two promising DG candidates- UNP and QCOM. Both are actually a little below my average price but offer great promise as long term DG holdings.
    I will add to a position until it reaches 4-5% of total portfolio, then I stop to avoid "concentration" in one position. I will actually start selling a position if it gets above 10%, as studies show your risk from a "single stock" (single comapny risk) goes up significantly when it is 10 plus percent of your portfolio.

  3. I follow your methodology as well. Large sums of money spread out and then pick your best idea. I do need to pay better attention to weights though. Like you said one of the goals of a DGI is to minimize risk.

  4. The method isn't mentioned here but what I have done since I initiated my DGI portfolio in 2008 is equal weight by income. I feel this method is less risky as the higher yielders comprise less $ content than the low yielders yet provide the same income.

  5. I'm very early in the accumulation phase so I don't feel it matters all that much to me, I'm still adding companies at a steady fixed monthly amount to my portfolio. However when I need to draw down income, I think I like the idea of weighting it based on income. I've seen some people do that on SeekingAlpha and although it makes sense to me, it is rare to hear it suggested or to find people that do it. It's basically the opposite of the second method, where the higher yield companies you'd have smaller positions in.

    The idea is that each position generates the same annual income, so if you have 20 companies in your portfolio and one of them stops paying dividends for some reason, your income only goes down by 1/20. As an income investor, my goal is to minimize the volatility in my income stream. As opposed to equal weighting based on dollar amount, which tries to minimize the volatility in my net worth without regard to how it impacts my dividend flow.

    Say your portfolio is 2 companies, one with 6% yield and one with 3% yield. To weight them evenly by income, you'd want 2/3 of your money in the 3% company and 1/3 in the 6%. Now if one company cancels their dividends you lose 50% of your income, but it doesn't matter which company it is that does so.

    For larger portfolio it can be annoying to do this weighting so I made a spreadsheet for the calculations that I'll share (was an exercise for a comment I left on a SeekingAlpha post a while back):

    So you can see the formulas to use.


Questions or comments? You can reach out to me at my website address name at gmail dot com.

Popular Posts