Seeking Alpha

Refining The 4% Withdrawal Rule

Includes: SPY
by: Investment Pancake

Retirees can live through horrific market conditions using a combination of dividends and principal for living expenses.

A spreadsheet tool that investors can use to see what mix of discretionary and non-discretionary spending survives historical market conditions.

By reducing fixed expenses and by cutting discretionary spending during bear markets, retirees may be able to withdraw far more than 4% of their monthly net worth on average.

Retirees who manage their own equity portfolios have three choices: live off dividend income, spend down a portion of principal, or do some combination of both. Either way, the big question is "how much can I afford to spend?"

Many advisors suggest that a retiree can safely spend an annual inflation adjusted amount equal to 3% or 4% of his or her initial equity portfolio. The problem with this static approach is that it ignores the fact that many retirees have a mixture of fixed costs as well as discretionary spending. In the real world, opening a brokerage account statement to find a bunch of red arrows can slap a wet blanket on a retiree's desire for indulgent discretionary spending, whereas a few months of heady gains can conjure up images of a brand new car or a nice vacation. With 147 years of history as any guide, this natural instinct to party when markets are high and eat stale crackers and cat food for dinner when the market is low can actually make for brilliant financial planning. That's interesting, isn't it? In almost all other cases, following your gut impulses when it comes to money, investing and saving is a recipe for disaster.

To try and answer the "how much can I afford to spend" question, I created a spreadsheet adapted from Professor Robert Shiller's CAPE spreadsheet that traces historical stock market data going back to the 1870s (which is available on Professor Robert Shiller's home page). You can find a link to my adapted version of his spreadsheet at the end of this article. In my article last year about immortal retirees, I explored the topic of how much a retiree could spend each year based entirely on dividend income. This time, I decided to look at a combination of dividend income AND capital gains.

First, I assumed that a hypothetical retiree has only one income source: a stock fund that tracks the S&P 500 (SPY). I assumed that the hypothetical retiree will spend a fixed amount each month taken from either monthly dividend income OR, to the extent that the dividends aren't sufficient, from principal. This is a non-discretionary amount that the retiree will spend regardless of market conditions.

Next, I assumed that the retiree will indulge in discretionary spending each month based entirely on how well the stock market performed in the prior month. For purposes of the spreadsheet, I set a rule that if the market is up over the last month, the retiree will spend the greater of some fixed dollar amount or a percentage of the last month's gains. You can use the spreadsheet yourself to experiment with different withdrawal percentages or discretionary fixed dollar withdrawals. Or, you can change the rule to withdraw principal when markets are up on a year over year basis, rather than a month over month basis. The key assumption is that the retiree follows her basic human instinct: when markets are up, discretionary spending goes up, and when market are down, the retiree tightens the old belt and only spends the bare minimum of required, non-discretionary expenses.

Here is an example of what the historical data shows. Assume a hypothetical retiree who retires in 1925. The stock market is booming and the road ahead looks very promising. The retiree has a portfolio worth about $1.5 million in today's dollars, fixed costs of $2,000 a month in today's dollars, and a desire to spend at least an additional $2,000 a month if the market is even slightly up from the month before. If the market is up sufficiently, the retiree will step up her discretionary spending beyond $2,000 and spend an amount equal to 25% of her portfolio's gains from the prior month.

How well would this person be doing after 30 years in retirement? Having survived the crash of 1929, the Great Depression, and World War II, the retiree would have a portfolio worth $1,526,011 in today's dollars, and would have been able to spend a total of $3,423,620 throughout her retirement. That comes to a total return of 325%. In fact, the retiree could stretch her retirement to the present day and end up with a portfolio worth $1,979,283 and total cumulative spending of $9,722,440 since 1925. That's a total return of 946%, and it turns out that the average spending rate for this retiree comes to 7% of her monthly net worth. This is dramatically higher spending than the 4% withdrawal rule, and assumes that she rides through the very worst market conditions in U.S. history.

Here is a chart from the spreadsheet tool that tracks the retiree's total spending and portfolio net worth from 1925 to the present.

What if this same retiree had much higher fixed costs? Assume the retiree had retired in 1925, but instead of needing to spend $2,000 a month in today's dollars, her monthly needs are $3,000. However, she would only want to spend an additional $1000 a month in discretionary spending if the market had really taken off the month before. It's the same combined amount of discretionary and optional spending as the previous example, but if you boost the fixed spending higher and the discretionary spending lower, the retiree would end up going bankrupt by 1986.

Are there any lessons to draw from the exercise? After experimenting with multiple scenarios and retirement dates, my conclusion is that if a retiree is willing to avoid discretionary spending when the market is down, then the historical data suggest that a static rule for withdrawing principal (such as the 3% rule or 4% rule) can be a spending floor rather than a spending ceiling. Even if a retiree had retired a few years before a 90% stock market plunge and a generation-length bear market, the retiree could spend 3% of her initial retirement portfolio each year on non-discretionary items, and an additional amount of up to 17% of her previous month's portfolio gains (if any). She'd still be solvent even after a 90-year retirement. The following chart illustrates the net worth and cumulative spending for this hypothetical retiree who follows a "fixed 3%, discretionary 17% of last month's gains" portfolio withdrawal plan.

For the sake of comparison, here is a chart showing the total spending and net worth of a hypothetical retiree who spends a fixed 2% of her initial portfolio net worth, and a variable amount equal to 20% of the prior month's gains (if any) from 1925 to the present.

The "fixed 2%, variable 20%" plan enables the retiree to spend far more ($16,802,000 from 1925 to the present) than the "fixed 3%, variable 17%" plan (which gives her $11,020,627 from 1925 to the present).

My conclusion is that over time, a retiree may be able to enjoy far higher overall consumption by having relatively low fixed costs and relatively high discretionary spending that the retiree adapts to market conditions. If my conclusion is valid, then it may be relevant to a retiree who is trying to decide whether or not to pay off her mortgage prior to retirement, or who is considering moving to a low cost jurisdiction. Most of all, it may indicate that there is at least one area in the personal finance arena where it pays to do what comes naturally when you open your monthly brokerage statement.

Here is a link to the spreadsheet. You can use it to enter in any portfolio amount, any fixed spending, and any discretionary spending you want to study from 1871 to the present. After you click the link, go to "file" and then "make copy." I encourage you to leave comments on any findings or conclusions you are able to draw from the data.

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 article, and the attached spreadsheet, are not investment advice and cannot be relied upon by any person for any reason whatsoever, other than entertainment.