Suppose that your plan is to retire and live forever. Suppose that your goal is not to maximize your wealth but to maximize your spending power without going bankrupt. Based on historical market data going back to 1871, how much could you safely withdraw each month from a portfolio that tracks the S&P 500?
The standard approach to answering this question is to look at a portfolio at a particular moment in time (such as a retirement date) and to derive a static amount, adjusted for inflation, that can be safely withdrawn from the portfolio without entirely depleting it. Some retirement calculators, for example, will base this static withdrawal amount on a percentage of portfolio assets determined as of a specific retirement date.
The drawback of this type of approach is that over very long periods of time (which you might see with a multi-generational trust or pension plan), the static withdrawal amount can become paltry in comparison to a portfolio's net worth and income-generating capacity. Vlae Kershner (another Seeking Alpha commentator) raised this point in the comment stream to my article published on Seeking Alpha last October, entitled "The Vampire's Retirement Plan." In that article, I found that a retired vampire with an infinite lifespan could survive the 1929 stock market crash by withdrawing a fixed amount (adjusted for inflation) equal to 80% of his portfolio dividends as of the retirement date. Over the years, the vampire's net worth becomes enormous compared to his spending, which prompted Mr. Kershner to ask a very apt question: at what point can the vampire enhance his standard of living?
This time, instead of looking only at a static withdrawal amount based on portfolio dividends as of a certain date, I used a two-tier approach. I used a static minimum withdrawal amount plus a variable "bonus" withdrawal amount equal to a percentage of future portfolio dividends. If the dividends exceed the minimum required withdrawal amount, the vampire will spend, save or reinvest a certain percentage of those excess dividends. If not, there will be no bonus withdrawal.
The specific approach was to take the data spreadsheet available to the public on Professor Robert Shiller's database at the Yale School of Management and to add variables for future spending, investing and saving rates for portfolio dividends in excess of an inflation-adjusted minimum required distribution amount. You can access this spreadsheet, download a copy for yourself, and then run as many scenarios as you wish to explore the impact of different retirement dates, portfolio size relative to minimum spending needs, and future dividend reinvestment rates, spending rates, and savings rates.
So with that background in mind, let's now take a step back through the halls of time. The year is 1871, and you have just arrived by ship to the New World. You are one of the undead. You've lived in Paris, London, the orient, the Carpathian mountains, and you have amassed a small fortune over the course of your long years. Now it is time to cash in.
You enjoy living like the petite royalty you once feasted upon and are content to spend a minimum of $20,000 a month (in today's dollars) on the various luxurious extravagances to which you have become accustomed. Based on Professor Shiller's data, that comes to about $1,030 a month in 1871 dollars. (In cell H3 of the spreadsheet, you can input any minimum withdrawal amount you wish in today's dollars and the spreadsheet will automatically translate that into 1871 dollars.)
In your Goyard steamer trunk, you have stacks of currency and gold coins worth $197,600 - or roughly $3,108,436 in today's dollars. That happens to be just enough for you to purchase 35,000 shares of an S&P 500 fund at prices prevailing in January of 1871 and still have $5,000 of cash left over. (In cell I3 of the spreadsheet, you can assign any number of portfolio shares you wish, and in cell T2, any amount of cash savings you desire.)
You plan to withdraw a minimum of $1,030 a month every month, and adjust that amount to account for inflation (or, in some cases, deflation). You will first look to your dividend income to cover your $1,030 monthly costs, but if your dividend income is insufficient, you will draw down your savings account. If your savings account runs dry, then you will make up the shortfall by selling shares of your S&P 500 funds at whatever price is then available.
However, if your dividend income in any month exceeds your minimum $1,030 amount, you'll spend 90% of that excess, put 5% back into your savings account and reinvest the remaining 5% back into more S&P 500 funds at then-prevailing prices. (You can change the proposed allocations between spending, saving and investing in cells M3, N3 and O3, respectively.) Together, the inflation-adjusted $1,030 baseline spending, plus the 90% "bonus" dividend spending, are what I define as your portfolio's "spending power" in column U of the spreadsheet. You only care about spending power. You are a libertine. You crave custom-tailored black suits and expensive, creepy-looking sunglasses. Your goal is to maximize this "spending power" figure over time. A really, really long period of time.
When you discuss the plan with your eager young broker, he seems dubious. He points out that your baseline monthly spending habit is equal to 7.72% of your entire net worth as of January 1, 1871. (See column W on the spreadsheet.) He points out that your monthly dividend income is only $758 - leaving you with a shortfall of nearly $272 as of your very first month of retirement. Will history bear out your broker's concerns?
The answer is that by 2016, your spending power will be nearly $109,000 per month, a mere 1.93% of your burgeoning $67,480,000 net worth. There will have been plenty of bumps along the road to get from 1871 to 2016, but it looks like you overcame these with a wide margin of safety. A picture says it all.
But in fact, looks are deceiving. Had you retired in 1871 with only 33,600 shares of the S&P 500 (as opposed to 35,000 shares) and $5,000 in your bank account, the spreadsheet shows that you would have been bankrupt by 1994. And by the same token, had you retired with 33,600 shares of the S&P 500 in 1871, but with a cushion of $5,250 in your bank account, by 2016 you would have a net worth of $7,840,256, and monthly spending power of $20,000 a month. It turns out that a tiny boost in cash reserves spares you from bankruptcy, but does little to enhance future spending power. The chart, below, says it all.
But now suppose that you retired in 1871 with $5,150 in the bank and 33,600 shares of the S&P 500, but instead of spending 90% of the excess "bonus" dividends, you spend only 50% of your excess dividends, put 10% into savings and reinvest 40% of the excess dividends back into the S&P 500. This modification to your plan would have provided you with a net worth of $66,004,932 by 2016, and spending power of $67,483 per month. In sum, the vampire's savings and investment behavior after retirement influences the portfolio outcome at least as much as the vampire's initial net worth and minimum required spending as of the retirement date. Here is a chart of the result.
This spreadsheet tool provides a better picture of portfolio spending power than a model based only on a static withdrawal amount, but it still has limitations. I ran multiple scenarios on the spreadsheet and found one very interesting pattern. Column W on the spreadsheet calculates the spending power of the portfolio as a percentage of the portfolio net worth. In the scenarios I ran, by the time the vampire's consumption levels are routinely in the area of 3% of his portfolio net worth, wealth begins to rapidly outgrow portfolio spending power. We see runaway wealth accumulation kick in once the vampire's spending is routinely in the area of 2% of his or her net worth. You can change the retirement year from 1871 to other years, and either increase or decrease future savings and investing rates, and still see the same pattern emerge. It seems that even a static "bonus" withdrawal rate based on future dividends will eventually understate the spending power of a portfolio after extremely long periods of time.
On a technical note, when you click the link to the spreadsheet, you will only be able to input your own variables if you save a new copy. Make sure to click "file," and "copy" on the upper left hand section of the screen. The screen will look like this:
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
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.
Additional disclosure: This is not investment advice and I am not an investment advisor.