Seeking Alpha

Dividend Growth Investing For Retirees

by: R. Paul Drake

DGI portfolios offer potential promise for retirees, to grow wealth or sustain spending.

I evaluated growing wealth and sustaining spending using Monte Carlo simulations.

Sequence of returns risks are significant, as is the effect of dividend cuts.

I compare the results for DGI portfolios based on the Dividend Aristocrats with results corresponding to S&P 500 index funds.

Dividend Growth Investing (DGI) has numerous advocates on Seeking Alpha. Many of them are focused on the long-term growth of dividend income, with the perspective of a working investor, steadily accumulating funds. In contrast to this perspective, the present article explores the relevance of DGI investing to retirees.

This topic is personally important to me. I am on the verge of obtaining full control of substantial funds whose use has been restricted to the mutual funds available in my employer’s 401k-like retirement plans. So the question of whether to place some of these funds into a DGI portfolio has personal relevance.

I evaluate the prospects of potential investments using Monte Carlo simulations, working with the historical mean and variability of the returns on those investments. There are other factors to consider as well, such as the rates of dividend cuts and of bankruptcies.

I introduced and discussed Monte Carlo simulations here. If you have let any investment advisor work up an evaluation for you, or used the web tools that are widely available, then you have almost certainly been shown results of a set of Monte Carlo simulations.

After some explanatory material, results are shown below for two cases relevant to current or near-term retirees. The first case is a growth-portfolio case. A newly retired investor places 500k 2019 dollars into a DGI portfolio for 20 years, planning to draw from these funds after that.

The second case is a steady-income case. A newly retired investor places 500k 2019 dollars into a DGI portfolio and withdraws some percentage of the real value of the initial portfolio each year to support retirement spending. For this case I looked at 30 years.

The Dividend Aristocrats as a DGI Example

The Dividend Aristocrats are companies that have increased the cash value of their dividends every year for 25 years. As Figure 1 illustrates, since 1990 the total return (portfolio value with dividends reinvested) of the Dividend Aristocrats has been significantly larger than that of the S&P500. The total, geometric, compound annual growth rate ("CAGR") of the Aristocrats has been 12%.

Total return of the S&P 500 Dividend Aristocrats index and of the S&P500 index since 1990, frequently shown by DGI advocates. Figure 1. Total return of the S&P 500 Dividend Aristocrats index and of the S&P500 index since 1990, frequently shown by DGI advocates. Source

In first and second articles, I used the statistics of the Dividend Aristocrats index to evaluate the performance and risks of a DGI portfolio for a working investor steadily accumulating funds. Some of the material here reproduces material from these articles, to provide context for the results. If you want the full details on the methods used in the modeling discussed here, read those articles. Table 1 shows the statistical properties of the Aristocrats used for the modeling.

Table 1: Statistical properties of the Dividend Aristocrats for modeling.

Statistical properties of the Dividend Aristocrats for modeling.

Table 1 also shows the statistics of stocks dropping off the list or being added to it. It is notable that the rate of dropping off the list among the Aristocrats is comparable to the rate of dividend cuts for the market as a whole, discussed here. Being able to sustain dividend increases for 25 years apparently does not mean that a company is significantly more able to avoid dropping off the list when things go wrong.

Dividend cuts among the Aristocrats in recent years have been historically low. One should expect this trend to end; it is very likely not to continue beyond the onset of the next recession.

Two remaining parameters are important. First, some fraction of the dividends is reinvested. For a retired investor, this might vary from 0% to 100%. This will depend on whether the DGI portfolio is being left alone to grow or is being used for current income.

The second parameter, "fraction recovered", relates to what happens when a stock drops out of the Dividend Aristocrats. I discuss this parameter at length in this article. It accounts for whatever losses (or gains) an investor experiences, relative to the annual value of the index at previous year end, when a dividend is cut so that the stock is removed from the index.

The Monte Carlo Modeling

The main question Monte Carlo simulations can address is what spread of outcomes will be likely for some investments, if the markets behave statistically as they have in the past. This, in particular, can help one understand sequence of returns risks.

For the modeling discussed here, the portfolio consists of 50 stocks, with new ones added when old ones drop off the list. The initial rate of return of these stocks is assigned according to the distribution of present yield. The dividends are all increased each year at the mean annual return for the historical Dividend Aristocrats.

The rate at which stocks drop off the list in each year is drawn from a distribution based on historical statistics. The rate of return for the portfolio in each year is drawn from a distribution based on historical statistics.

Results for a DGI Growth Portfolio

The first case shown here assumes that a newly retired investor sequesters $500k into a growth portfolio. The investor reinvests 100% of the dividends, replaces stocks whose dividends have been cut with other stocks, and does not withdraw any funds from the portfolio.

This strategy, combined with a focus on dividend income, enables the investor to benefit from the strong psychological benefits of taking a DGI approach. If one’s main focus is growing the income and not total return, one will be far less tempted to sell out during market downturns.

Figure 2 shows how the dividend income generated by this portfolio increases with time. The vertical bars show the range from the 20th to the 80th percentile. The range of outcomes on this plot reflects the range of dividend yields present in the initial portfolio.

Dividend income from the model portfolio of Dividend Aristocrats. This begins with $500k of 2019 dollars. Figure 2. Dividend income from the model portfolio of Dividend Aristocrats. This begins with $500k of 2019 dollars. The model used 1,000 simulations for each case. The upper and lower curve correspond, respectively, to a fraction recovered of 100% and 50%. Calculations by author.

The results shown in Figure 2 are excellent. The mean yield on cost, if one has avoided any losses when stocks drop off the Aristocrats list, is 14%. Even if one does quite poorly and loses half the capital invested in stocks that drop off, one has a yield on cost of 10%. And this is in real dollars, 20 years later.

Beyond that, one can sensibly anticipate that the dividend income will more than keep pace with inflation going forward. On the assumption that the investment universe will continue to allow Dividend Aristocrats to emerge and flourish, it would be hard to argue against this as a viable approach.

Let’s look also at what has happened to the total portfolio value. Figure 3 shows this. Again the vertical bars show the range of results from 20th to 80th percentile.

Total value of the model portfolio of Dividend Aristocrats at the end of 20 years. Figure 3. Total value of the model portfolio of Dividend Aristocrats at the end of 20 years. This begins with $500k of 2019 dollars. The model used 1,000 simulations for each data point. The red point shows the corresponding results based on historical statistics of the S&P500. Calculations by author.

One can see that so long as one recovers more than 50% of the invested capital when a stock drops off the list, the 20th percentile limit of the results does better than the corresponding limit of the S&P500 portfolio. In one recovers more than 80%, then the mean result is higher.

Notably, the value of the portfolio probably reaches 5 or more times the initial invested capital. If, rather than spending the dividends, one were instead to withdraw 4% to 6% of the total each year to provide income, then the likely income would exceed $100k to $150k. That represents 20% to 30% of the initial investment, in each year.

The breadth of the distribution of portfolio values, shown by the length of the vertical bars, reflects the impact of sequence of returns risk. This is larger than the difference between the outcome for an S&P 500 index fund and the results of the DGI portfolio.

Results for Spending from a DGI Portfolio

For this case I assumed that the investor, upon retirement in 2019, places $500k into a DGI portfolio and then withdraws the same fraction of that $500k, in 2019 dollars, every year for 30 years. Note that taking this approach is one alternative to putting those funds into a bond ladder or buying an annuity, with the potential to support much larger spending rates.

Portfolio value after 30 years, in 2019 dollars, vs annual rate of withdrawal. The DGI case is shown in black. Results for an S&P 500 index fund are shown in red. Figure 4. Portfolio value after 30 years, in 2019 dollars, vs annual rate of withdrawal. The DGI case is shown in black. Results for an S&P 500 index fund are shown in red. Calculations by author.

Figure 4 shows the results, for the case of a fraction recovered of 90%. As one can tell from Figure 3, the DGI mean values (the black points) will move up or down as the fraction recovered increases or decreases.

In Figure 4, the vertical bars show the 20th to 80th percentile values of the portfolio. In the highest-withdrawal cases, the portfolio may have been exhausted more than 20% of the time, so that this lower limit is stuck at zero.

Over 30 years, the incremental gain associated with the Dividend Aristocrats does add up, so that the mean portfolio value is larger than that for the S&P 500 portfolio. The advantage remains modest; the spread of results produced by sequence of returns risk is larger than the increase in mean value.

The behavior of the 20th percentile limit may be more significant in practice. One sees that, at the 20th percentile limit, the real value of the DGI portfolio remains above the initial invested capital up to a withdrawal rate of 8%.

I believe that this behavior reflects not only the larger total return of the Dividend Aristocrats, as compared to that of the S&P 500, but also the fact that the dividends decrease the amount of portfolio depletion in down years. A test using the total return of the Aristocrats but no dividend yield produced results consistent with this belief. It shows broader sequence of returns variations and lower 20th percentile points for the no-dividends case.

Portfolio and Spending Implications

I end up with opposing reactions to the two cases analyzed above.

On the 20-year timescale over which one might grow some dedicated funds, in order to support the later years of retirement, the advantage of the Dividend Aristocrats over an S&P 500 index fund is not large compared with the sequence-of-returns variations. In addition, for myself, I don’t have any confidence that my own stock picking could do better than choosing the Dividend Aristocrats.

As an alternative to using an S&P 500 index fund, there are several ETFs oriented to dividend aristocrats by some definition. Examples include (NOBL), (REGL), (WDIV), and (SDY). They have low expense ratios. It might make sense to put funds there rather than to go to the effort of doing it yourself.

It would still be worth taking a DGI approach if this would prevent one from selling into bear markets. As an alternative, it might make sense to pay a financial advisor to run a constrained, Dividend Aristocrat portfolio for you and to talk you out of selling during bear markets.

In contrast to the growth-portfolio case, the DGI approach using Dividend Aristocrats appears to have real advantages while one is spending from the portfolio, as we saw in Figure 4. The model suggests that you can withdraw 8% of the initial real value of the portfolio every year, and still have an 80% chance that its final value will be larger than its initial value.

My reaction is that this would be a great plan. Do it for five years without looking at the portfolio. Assign 1% to a financial advisor and take 7%. Go have fun. Then assess in 5 years what to do for the next five years. The odds are that you will be able to spend more rather than less.

The ability of the dividends to resist portfolio collapse is a major positive feature of taking a dividend-oriented approach. It is natural to wonder whether there might be a benefit of going up in yield from the low levels of the Aristocrats. I found some data that enables some modeling of this. Stay tuned for a next article.

The bigger risk here than sequence of returns risk is that of black swans. If events change the return of the stock market in enduring and negative ways, this will hurt you. This is why every investor needs diversification into other sorts of assets and other countries.

The classic diversifying asset classes are international stocks, both developed country and emerging market. My personal holdings in these areas include the TIAA-CREF funds (TCIEX) and (TEMLX). Until recently, I also held (FSPSX) and (FEDDX) with Fidelity. My holdings also include the Fidelity Low Priced Stock Fund (FLPSX). It is value-oriented and has not excelled during the present bull market for growth stocks. But its lifetime (30 year) average return has been 13%. I expect strong outperformance when the S&P 500 next suffers a real bear market.

The additional possibilities for diversification include real estate, private equity, bonds, MLPs, and precious metals. For the long term, I am not a fan of bonds in the present phase of the interest rate cycle. I personally rely on High Yield Landlord to help me consider REIT investments and some MLPs.

Disclosure: I am/we are long (TCIEX) (TEMLX) (FLPSX). I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.