The 4% rule goes something like this: place your funds in blue-chip stocks and in the first year of retirement, take out 4% of the total balance and then in each year, adjust that percent up or down for inflation.

So, does it work?

The first part of this is this: is 4% the correct amount? Will it meet your needs? See discussion of this at this post.

Now, assuming it is the correct amount, does it work? By this I mean, will it last through a retirement? There are a few variables.

First, how many years? To some extent, people can generally control when they retire; it’s the longevity that is a guessing game. But, something has to be decided, so I looked to the Social Security mortality tables, which tell us that at age 55, a male is expected to live another 25 years, a female, 28. by age 65, those numbers are 17 years for the males and 20 for the females. This is why people generally use a 30-year retirement period to compute needs.

That may be a little aggressive because these are averages, but it is a starting point.

To make an analysis of the adequacy of the 4% rules, I did the following:

- Gathered the month end closing numbers of the S & P 500 from January 1, 1950 to December 31, 2000. (The theory here is that the country and the financial markets went through a lot during this period, probably some of the most tumultuous years we have had and that using these returns covers what might happen in the future. People will disagree with this, and if so, I have no quarrel with them. The idea is to replace pure theory with some form of reality.) I wanted 600 periods, and me, being the CPA and all, ended up with 612, which I did not notice until after all the numbers were crunched and then was not going to go back and correct for this, so, I ended up with 612 periods.
- Computed the monthly return for each of the 612 periods. For example, the S & P was 17.05 as January 3, 1950 and was 17.22 at February 1, 1950, a change of 0.17, or 0.997%, a little less than 1%. After the end of the 612 periods, I started over with period one’s gain or loss.
- Using as a proxy beginning amount of $1 million, the initial monthly draw was computed as $3,333 (1,000,000 X4%/12). This was inflated at 3% per year, so that the 13th draw was $3,433.
- Set up a table, beginning as of January 1, 1950 with $1 million, added the market gain or subtracted the market loss and deducted the withdrawal of 4% as adjusted for inflation. This was repeated for each month since January 1950 until December 2000.
- Looked at the results to see how many years the funds lasted.

I used a 40-year retirement period as the base. Out of the 612 periods, 413 made it this far, or 67% of the time. What this says is that 2 out of 3 times, the funds would last this long. Whether 2 out of 3 is bad or not, is for each individual to decide.

At 35 years, the funds made it 74% of the time. At 30 years—the old stand-by period—the funds were sufficient 81% of the time. And at 25 years, they were sufficient 92% of the time. 13 of the tests did not make it 20 years, which is about 2%.

The average rate of return of the S & P during the 612 month period was 7.2%.

A couple of tweaks. Starting at 3.5%, the funds lasted 40 years 84% of the time and at 3%, the lasted 40 years 96% of the time. To pull out the same $3,333 per month at 3% would require a starting fund of $1.333 million.

What this tells is that timing of the retirement makes a huge difference. If immediately after you start drawing, there is a bull market for a few years, you will become very wealthy. If it is down for several years, the funds may deplete before you die.

What this tells me is that the 4% rule may have more risk than many will want to absorb.

If your brain hurts from reading this, just imagine the mind-numbing effect of the computations themselves.

As I said in the first part of this discussion, talk to a professional.

None of this information is advice. It does not take into account any of the hundred things that will need to be considered. I am a Certified Public Accountant and I consult with experts in these matters. You should also, because without a trained person assisting you (me), it is all too easy to overlook something. And what is right for one person just may be flat wrong for their brother or neighbor.

So, think about these things and start planning early, and by early, I mean at least 30 years before retirement, and talk to a professional, ideally a Certified Financial Planner, who will charge you a fee, but her advice could be the difference in annual (or bi-annual) trips in your golden years to Paris, France or Paris, Arkansas (and I am not knocking the Arkansas version, but is does not have the Louvre and all).