Why I was a disbeliever of the 4% retirement rule!

For those of you not familiar with the 4% rule, it is a general rule of thumb used by financial professionals (or commentators) on a safe annual withdrawal rate during retirement.  The general idea is if you never withdraw more than 4% (I believe the exact rate is 4.2%), in any given year, you will be able to safely withdraw from your savings for the remainder of your life.  Several studies have been done on this, but William Bengen is given the original credit for this rule.  Having reviewed several of the studies there appears to be pretty strong empirical support for 4% retirement rule.

What I have seen though is the 4% rule interrupted in a slightly different way.  There are several early retirement bloggers and authors, that use the 4% rule in figuring out how much you need to save.  They also call it the “25 multiple rule”.  It is a small but important distinction.  Let’s look at an example, say you need $40,000 to live, by this rule you need to save $1 million dollars ($40K times 25 which also equals 4% of $1 million).  If you need $80,000 in annual income, then you need to save $2.0 million.  According to them, once you have saved this amount you are done.  You can retire and live on this amount for the rest of your life!  Simple right?

I have for quite some time believed this logic was flawed, as the original 4% rule discussed safe annual withdrawals, as opposed to a retirement goal amount.  My concern was by setting that amount as a specific goal, it did not give the early retiree the flexibility to weather a down market.  I believe and still do that if you say I need $40K to live on, you likely won’t have the ability to cut that to $30K in a pinch.  Thus in any given year, you would withdraw vastly more than 4%.  My logic on that is pretty sound if you look at market returns during historical recessions.  Too me this is too overly simplistic and not factoring in long periods of poor market returns.  For the record, I am not the only one that believes this.

The question I wanted to know is could the early retirees recover enough to still be able to say the 4% rule is sound for setting the retirement goal amount?

So being the analytical nerd that I am, I decided to look at historical market returns and see if you could set a dollar withdrawal rate at beginning of retirement.  Then continue to withdraw this rate plus inflation and ride out the market ups and downs for an extended period of time.

To give you an idea of the information I analyzed I looked at the following:

  • Used historical total returns of S&P 500 going back to 1928 through YTD 2017. As a disclosure, I saw some minor variation in different sources for this data, but the difference was 1 or 2 bps (0.01% or 0.02%) and only in certain years. Ultimately that small variation had no material impact on the results.  And yes, this factor in dividends.
  • I looked at both a $40,000 annual draw rate on $1 million and $80,000 annual draw rate on $2 million to see if the size of initial starting portfolio plays a role. There was a difference, but it was minor.
  • Each annual withdrawal was withdrawn at the beginning of the year. Now you might say that is being too conservative, but I think that is more likely or more practical.  You are not going to sell stock holdings today to pay your weekly grocery bill.  You most likely will sell stock ahead of time to cover future bills.  I think planning a year ahead is not only reasonable but practical.
  • The retirement start year was looked at during every year from 1928 through 2008.
  • Factor in annual inflation by using the 10-year average inflation rate of 3.22%, which is close to historical average of approximately 3.20% (link).
  • In addition to using historical average inflation, I also looked at the period of high inflation to see how much of an impact that would have on the outcome.
  • My result of success was getting at least 50 years of withdrawals while factoring for inflation. Obviously, for the later years, I do not have sufficient time to see if that is successful but looked at the impact of the remaining balance and likely hood you could keep going with historical average returns.
  • I did not factor in taxes or tax planning. That would entail a lot of assumption and you should already be factoring taxes into your annual budget.  My focus was how market fluctuation affected the reliability of your income stream.

Okay, I know some will take issue with me just using the S&P 500, but in my opinion, it is the best long-term data available.  I do not remotely believe the Dow Jones is a better alternative, but you could argue that the S&P does not factor in foreign stocks.  It isn’t perfect, but nothing ever is.


I am very much the type of guy that just rips the band-aid off.  So what did the results tell me?

Well, I was wrong (don’t tell my wife).  The 4% rule is actually a pretty good rule of thumb to set your retirement goal too.  Now I say good, not perfect and it did not work in all time periods.  This method did not work if you started your early retirement in years 1928 to 1930 and it was questionable in the years 1999 & 2000, but it did appear to still work if you started your retirement in 2008.  The strong market recovery since 2008 help significantly.

What about high inflation rate?

Inflation can eat away at your purchasing power and it seems logical to look at a higher rate of inflation time period.  Now I am sure there are some out there that will say those days are over and we should be more concern about deflation.  Historically, we have had inflation over 5% in four different decades in the last 100 years.  Given the mass amount of monetizing easing we have had in this country, we should at least consider the possibility.

For this analysis, I looked at the period of 1970 through 1990, which had an average annual inflation rate of around 6.50% (round up).  Using the rate of returns for that period, it appears even at the higher inflation rate the funds would last for that entire period.  Even if you started in 1973, which was the worst market year during that period, it still worked.  Now if inflation continued you would eventually run out of money, but inflation does have an impact on market returns, so looking beyond 1990 isn’t fair relative to the market return.

What is my take away from this?

The 4% rule is a very good starting point, but it is a starting point not an absolute.  There are factors that can easily affect the success of your planning.  This simple rule doesn’t consider every possible variable.  Some of the things you should ask yourself:

  • How reliable or realistic is your starting budget to begin with? Did you actually track your spending over a year or try living on your projected withdrawal amount?
  • Are you making big assumptions of not getting this or paying for that, such as college for your kids?
  • Do you have high variable costs that could drastically change your budget?
  • How well can you handle an unexpected expense outside of your annual budget? If you retire at 40, living another 50 years isn’t unrealistic and a lot can happen in 50 years.
  • How diversified is your portfolio? Think of the Dot.com bust.  If I used the Nasdaq rate of return this would not have worked.
  • Are these retirement funds your sole source of income? If you have some other income stream that will give you some flexibility over your withdrawal rate.  The more flexibility you have the greater certainty of success.
  • Do you have insurance to cover the big items, like medical or liability?

One of the things I noticed, was that the two time periods it did not work involved three years of back to back declining market.  If you had enough flexibility to stop withdrawing money for a given 12-month period, that would have allowed for the  great probability of success during all time period.

I can say I am more supportive of the 4% rule, but there should be an asterisk that indicates that the ultimate success depends on building sufficient cushion and flexibility into your budgeted assumptions.

Would love to hear your opinion, so please leave a comment.

Thank you




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s