Here's a Third Way to Determine When You Have Enough to Retire

by: Investment Pancake
Several Seeking Alpha authors have recently posted thought-provoking articles about retirement, and how much money a person needs to do it. Two excellent articles available at Seeking Alpha include With the Right Dividend Stocks, How Much Money Would you Really Need to Retire, by Dividend Growth Investor, and Will High Dividend Stocks Help You Retire Comfortably?, by Geoff Considine. The debate boils down to which of two analytic models best gauges a person's retirement preparedness.

Model number one assumes that a person can draw down 4% of his nest egg with relatively low risk of outliving his money. Some studies suggest a 3% withdrawal figure is more appropriate, while others suggest a person could withdraw up to 5% of the portfolio each year without undue risks of bankruptcy.

While ample statistical evidence supports the use of the 4% model, there's a major flaw with the approach in that it assumes a constant withdrawal rate under which a retiree will sell assets into weak markets in order to fund his consumption. Put differently, the 4% model assumes the very worst sort of investment behavior on the part of the retiree: forced selling during bear markets. In fact, people tend to spend less during bear markets, so assuming a constant 4% withdrawal rate doesn't necessarily comport with actual human behavior. Moreover, keeping a steady 4% withdrawal assumption will skew the results of a Monte Carlo simulation towards increased bankruptcy risk for a retiree.

The second model looks less at a person's net worth, and more at how much income a person's portfolio tends to generate. Some suggest that if you can build a portfolio of assets that produce a somewhat steady rate of income that tends to increase with time (for instance, dividend income from companies that typically raise their dividends each year), you can simply withdraw all of that income each year to live on without fear of going broke, and with less concern about inflation. The main drawback to this second approach is that in fact, companies can and often do slash dividend payments. Those who relied on bank dividends to fund retirement plans created in 2007 would be able to attest to the risks of using the second model.

I suggest a third approach would make more sense for those who are trying to peg a net worth upon which they could retire. My approach would work as follows:

First, as a baseline assumption, assume a retiree will own a portfolio that consists (in part) of dividend-paying equity investments, including shares of companies with strong records of dividend raises that keep pace with, or exceed, the rate of inflation. The portion of the portfolio comprised of such assets will depend on the retiree's risk tolerance, and the retiree should project that his spending should ultimately be funded almost entirely through the income generated from this portion of the portfolio.

Second, assume the portfolio also has some "income insurance" - which is to say, the portfolio will include assets that tend to go up in value during periods of risk aversion, which is when companies are most prone to slashing dividends. This "insurance" section of the portfolio will largely consist of longer-term, ultra low risk government bonds. During periods when companies are slashing dividends and the retiree's income is falling, the retiree will plan to use some of the capital gains on his low-risk bond portfolio as an income replacement, hopefully offsetting the diminished dividend payments. Otherwise, the retiree should plan on reinvesting all interest and capital gains on the low-risk bonds back into more low risk bonds, and in this manner, ensure that his income insurance policy grows at a rate commensurate with inflation.

Third, the retiree should break down his costs based on two separate scenarios: (1) how much would he like to spend when times are good, and (2) how much would he absolutely need to spend during lean times. Ideally, this former figure should be linked to the projected yield on the retiree’s income portfolio, and this latter amount should be closely linked to the projected gains on low-risk bonds in the retiree's portfolio.

Here's a case study to illustrate how I might use the approach. First, assume Sandra would like to live on $120,000 a year, adjusted for inflation. In a pinch, she could get by on $80,000, covering her health care costs, mortgage, food and electricity, but forgoing vacations, a new car, or trips to restaurants. Let’s also assume Sandra is very risk tolerant, and doesn’t care much about what her net worth is, so long as quarterly checks come in that are large enough to cover her costs.
Sandra finds that she can invest in various equity ETFs that consist of companies with strong track records for increasing dividends. For instance, the SPDR S&P Dividend ETF (NYSEARCA:SDY) produces a yield of 3.36. Powershares International Dividend Achievers ETF (NASDAQ:PID) produces a more modest yield of 3.06. Sandra also finds a number of master limited partnership funds producing tax-advantaged yields of around 6%, including Tortoise Energy Capital Corporation (NYSE:TYY) and Kayne Anderson MLP Investment Corporation (NYSE:KYN).
Sandra also finds some business development corporations with extremely high dividend payments but less track record when it comes to dividend increases. Apollo Investment Corporation (NASDAQ:AINV) offers a 10% yield, and BlackRock Kelso Capital (NASDAQ:BKCC) offers an 11% yield. With these assets, Sandra is able to construct a portfolio that churns out cash at a yield of about 6% a year, and which is likely to go up over time as the funds and companies she owns raise their dividends.
The first thing Sandra needs to fund her most comfortable retirement is a $2,000,000 portfolio of income producing, risky assets. Second, Sandra notes that dividends on her very risky portfolio could get slashed by 30% or 40% during a protracted bear market. She also assumes that if that were to happen, she’d tighten her belt and get by on $80,000 a year, at least until the dividends on her risky portfolio went back up again.
But Sandra is also a pessimistic person, and wants to figure that in a doomsday scenario, dividends on her portfolio would get slashed by 70%, which would leave her with cash distributions of only $36,000 a year, a $44,000 shortfall in terms of what she absolutely needs to survive.
However, Sandra notes that iShares Barclays’ 20 year plus Treasury Bond ETF (NYSEARCA:TLT) spiked about 30% during the great crash of late 2008. Sandra figures that during a lengthy spell of risk aversion, if dividends were getting slashed on her portfolio companies, and if she owned some shares of TLT, those TLT shares could rally by 20% during that period. To make up the $44,000 shortfall in her income that she might suffer if her companies and funds all slashed their dividend payments by 70%, Sandra would need over $200,000 worth of TLT to “insure” against this risk. So far, Sandra needs a portfolio of $2,200,000 - $2m of which consists of dividend paying equities, $200,000 consists of long term Treasuries or Treasury funds.
Sandra doesn’t completely buy the idea that Treasuries will absolutely have to rally during periods when dividends on her portfolio companies are getting slashed, so she decides that she’d like the ultimate insurance policy against periods of falling dividends – cash. Two years' worth, or roughly $90,000, should make up the projected shortfall she might suffer if her income portfolio sputters by 70%.
Using this approach, Sandra needs a portfolio of about $2,300,000 to fund a retirement that doesn’t depend on depleting her principal. She can spend about 5% of the initial face value of her assets, which is higher than the 4% rule, but less than the 6% yield on her income portfolio.
These numbers would change if Sandra was a more conservative investor. For instance, suppose Sandra is only prepared to invest 60% of her assets into equities, and wants a significant “what if all these assumptions go wrong” cushion. If so, Sandra might need a portfolio closer to $3,300,000 today to fund the retirement of her dreams, which puts her closer to the 4% withdrawal rule.
The benefit of this approach is that it incorporates some flexibility, which the 4% rule does not, regarding how a retiree might actually spend during both lean and flush periods. The approach also accounts for the fact that dividend payments get slashed at times. In effect, it steers a course between the two more frequently used retirement planning models.

Disclosure: Author holds long positions in TYY, KYN, SDY, PID, BKCC and AINV