Test-Driving an Income-Centric Retirement Portfolio

Much has been made of the so-called "war on savers"--the fact that the Fed's zero interest-rate policy has shoved conservative investors out on the risk spectrum. As yields on CDs and money market funds have shrunk, investors in search of a positive real yield have embraced all manner of higher-risk/higher-yielding securities. 

To test how the low-yield environment of the past decade and a half would have affected retired investors attempting to live off income from their portfolios, we ran a simulation along the lines of the retirement-portfolio "stress tests" we featured last week and the week before that. The simulations illustrate three of the key ways that retired investors could generate cash flow from their portfolios: pure total return, pure income, or a combination of the two. This week, we'll look at the pure income strategy.

Stress Tests in Review
The first simulation used what I've called a "strict constructionist" total-return approach, meaning that we reinvested all dividend and capital gains distributions back into the portfolio. We employed a cash bucket--part of the bucket approach to retirement-portfolio planning--to fund income needs each year, then re-filled that cash bucket each year with proceeds from an annual rebalancing program. To see details on that simulation--including spreadsheets that show how the rebalancing would have worked--click here. 

For the second simulation, we used a strategy that blends an income-centric approach (that is, spending dividend and income distributions in the year in which they're produced) with a total-return-oriented rebalancing program. When investments paid out income, we used them to cover current spending needs; but if those income distributions were insufficient, we looked to rebalancing proceeds to make up the shortfall. We rebalanced the portfolio annually and withdrew from the most highly appreciated positions. In contrast with the strict constructionist total-return approach, this simulation didn't employ a cash bucket. You can see the details on the blended simulation here. 

The Income Portfolio and Its Ground Rules
For our final simulation, we took a look at how a pure income-focused strategy would have fared over the same time period--2000-13. We stuck with a 50% equity/50% bond asset allocation, to help ensure that results weren't significantly distorted by asset-allocation variances. The specific holdings are different from the holdings in the first two simulations, however. Whereas the first two simulations employed the same fund holdings that appear in my aggressive bucket portfolio (with a few modifications to address the fact that some of the holdings had inception dates after 2000), we tweaked the holdings in the income-centric portfolio to emphasize current income. Our core U.S.-equity position in the income simulation, for example, is  Vanguard Equity-Income (VEIPX), whereas the other two simulations employed the lower-yielding  T. Rowe Price Equity-Income (PRFDX) and a total stock market index fund for U.S.-stock exposure. 

Note that we didn't go too far out on a limb in search of current income, even if it might have nudged up yield. We stuck with income-producing funds that earn high ratings from our analysts because they have strong records of balancing current income with total returns and risk controls. For simplicity's sake, we used funds rather than individual income-producing stocks and bonds. Although income-focused investors often prefer individual securities over funds to keep fund-management fees from dragging on their yields, we employed very low-cost funds to help mitigate that problem. (We had to pay a bit more for  Loomis Sayles Bond (LSBRX), which also appears in our other portfolios, because there's not a good lower-cost substitute.) 

Our cash-flow production and portfolio-maintenance regimen was straightforward. We harvested current income in the year in which it was distributed and used it for spending money, while reinvesting capital gains. Rather than seeking a steady inflation-adjusted payout from the portfolio--as is the case with investors who employ the 4% rule--we let income fluctuate along with our portfolio's organic yield. We did not conduct rebalancing. 

Of course, this is an extremely naïve approach. Some income-centric investors might aim to steady the portfolio's income production by venturing into higher-yielding investments when the existing holdings' yields drop, for example. In addition, income-centric investors might also engage in rebalancing in an effort to improve their portfolios' risk/return characteristics--focusing on income and employing a rebalancing program needn't be mutually exclusive. 

A Lumpy Income Pattern
As with the previous two simulations, our naïve income-centric approach delivered an ending portfolio value comfortably above its starting value, even though the time period studied--2000 through 2013--encompassed two major bear markets. One of the reasons is that the time period examined ended on an up note, as 2013 was an exceptionally strong one for equities. However, the income-centric portfolio scheme's ending value fell between that of the pure total-return approach and the hybrid approach (which involved taking income from the portfolio while also harvesting rebalancing proceeds); over the time period examined, the hybrid approach generated the highest return of the three methods. 

However, when cash flows from each portfolio scheme are factored in, as well as the portfolios' ending balances, both the total-return and hybrid approaches bested our naïve income-centric portfolio. With the first two simulations, we employed the 4% rule, extracting 4% of our portfolio's balance in year 1 of retirement and then inflation-adjusting that amount thereafter. Our hypothetical investor in those two simulations would have withdrawn $1,010,858 from the portfolio from 2000 through 2013, whereas the income-only simulation produced a total income stream of $962,974--almost $50,000 less. Additionally, the payouts under the income-centric approach were uneven and not adjusted for inflation; they were as high as $77,243 in 2007 and as low as $60,671 in 2002. 

As noted earlier, an income-centric retiree might take steps to help smooth income by venturing into higher-yielding securities in lean years, but doing so would likely heighten the portfolio's risk level, too. Moreover, this portfolio didn't benefit from rebalancing as did the portfolios in the previous two simulations, though an income-centric investor could easily rebalance, too. In both of the previous simulations, the portfolios benefited from adding to equities following their drops in the early '00s and again in 2008. 

It's also worth noting that not all of the divergences in performance and cash flows between this simulation and the previous two can be chalked up to portfolio strategy: Individual holdings selection and asset allocation played a role, too. For example, for the income portfolio we employed the  Vanguard International Value (VTRIX) in place of  Harbor International (HAINX) in the first two simulations because the former typically has had a higher yield; the Harbor fund outperformed the Vanguard fund over the time period studied. Moreover, the income-only portfolio had a slightly different asset allocation mix--it didn't include Treasury Inflation-Protected Securities or commodities, for example. 

That said, the simulation does illustrate some of the key trade-offs that can accompany an income-centric approach. Reassuringly, the income-focused investor was able to draw almost as much income in 2008's catastrophic market environment as she was in 2007, even though the portfolio's value had dropped. But the simulation also illustrates that income-producing investments, especially bonds, can see their yields rise and fall based on the prevailing interest-rate environment. That means that the income-centric retired investor must be content to make due on a smaller sum, venture into higher-yielding securities to help plump up the portfolio's payout, or consider harvesting rebalancing proceeds to steady the portfolio's cash-flow production on a year-to-year basis.

For a detailed look at the year-by-year results of our stress test, you can download this spreadsheet (Microsoft Excel required) or this PDF.

See More Articles by Christine Benz