Monday, December 14, 2009

Modifying the 4% Rule

What amount can an investor, Ian, withdraw safely each year from a portfolio 50% in stocks and 50% in bonds? Suppose that Ian wants to fix some dollar amount and bump it up by inflation each year, and wants a high probability of not outliving the money. According to a 1994 paper by William Bengen, the answer if Ian is between 60 and 65 years old is about 4% of his starting portfolio size. This has come to be called the 4% rule.

So, starting with $750,000, Ian should be able to withdraw $30,000 the first year, and then bump up the withdrawal amount by inflation each year. With reasonable probability, Ian won’t run out of money during his remaining lifetime, according to Bengen.

Patrick at A Loonie Saved noted a logical inconsistency with this 4% rule. Two investors in exactly the same situation might receive different advice. Let’s illustrate this with an example.

Suppose that in the first year of Ian’s retirement stocks performed very poorly. Between the $30,000 worth of withdrawals and the portfolio losses, Ian has $600,000 left today. Assuming inflation is 3%, Ian plans to continue with the 4% of starting portfolio rule and bump up his withdrawal amount to $30,900 for the upcoming year.

A second investor, Sam, is the same age as Ian, but is just retiring today. Sam’s portfolio is also worth $600,000. However, when Sam follows the 4% rule starting now, he will only withdraw $24,000 during the upcoming year. How can it make any sense that Ian and Sam are the same age and have the same amount of money, but should withdraw very different amounts during the upcoming year?

This inconsistency caused me to modify the 4% rule for my own use. When I get to full retirement, I plan to target withdrawing 4% of whatever my portfolio is worth each year. So, if my portfolio drops in value from one year to the next, my income for the next year will drop as well.

This kind of uncertainty may not be for everyone. I’m quite comfortable making lifestyle changes to increase or decrease spending. Many people can’t do this well. On the positive side, I won’t ever run out of money this way. If my portfolio happens to perform well, I will get to enjoy rising income.

The approach I plan to use may be forced upon some investors even if they planned to follow the original 4% rule. The poor performance of Ian’s portfolio in his first year of retirement greatly increases the risk that he will outlast his money if he doesn’t cut back on withdrawals. If he sticks to the original 4% rule for 5 or 10 years, it may become very obvious that he is running out of money fast.


  1. On the one extreme Ian could invest all his money in stocks and withdraw say 4% of the balance every year, indexed to inflation. The pro of this approach is he'll likely see his retirement income grow more than inflation. The con is there is a risk his income might not keep pace with inflation - this would occur if returns were poor in the first years of his retirement.

    On another extreme, Ian could use all of his savings to purchase say a bond ladder of government inflation-indexed bonds. His withdrawal rate will depend on interest rates and how long he wants the money to last. Say he wants the money to last until age 96, around 10 years after his life expectancy, and inflation-indexed interest rates are averaging 1.5%. Then he will be able to secure an inflation adjusted income stream of around 4% of his initial retirement savings. The pro of this approach is the certainty of his income, assuming he doesn't outlive it. The con is he is giving up the pro from the all-stock scenario, where he would likely see his retirement income grow more than inflation.

    I don't think either approach is wrong, it depends on your priorities and risk tolerance. For me, I would probably want some blend of stocks and bonds in retirement (maybe 50/50 or 40/60), noting that I might need to adjust my retirement income either up or down, depending on how the markets perform.

  2. Blitzer68: I agree that different portfolio allocations can suit different people. It all depends on how much you value extra income and how painful an income reduction is. However, where I think people can go wrong is in doggedly basing withdrawals on the initial portfolio value instead of adjusting income down in the face of poor portfolio performance.

  3. Yes, I agree. Thanks for bringing up the topic, it's an important one.

  4. Very useful analysis. I wish more were written about these types of calculations. I think the number of variables (asset allocation, inflation, interest rates, stock market performance) can scare a lot of us away from even looking at the numbers. But look at them we must for all of us will reach retirement some day.

  5. Thanks for the link Michael. Your remarks about lifestyle changes reminded me that I have made exactly these kinds of lifestyle changes in the past. (eg. I once left a well-paying job to finish my degree.) I've taken it for granted that a reliable income stream is a must, but perhaps not everyone needs that level of certainty enough to pay the price for it.

  6. I think of the 4% rule as more of a planning tool instead of an actual plan.

    I would find it difficult to sit back and watch things go up and down without making some adjustments to my withdrawals.

    My tendency would be to squirrel away some of the excess in the good years so I would have more to spend in the lean years.

  7. 2 Cents: Not to be too much of a downer, but some of us won't reach retirement, either because we choose to work late in life or we die young. However, the vast majority of us need to plan based on the assumption that we will retire one day.

    Hazy: I agree that I would probably tend to spend less than 4% in good years to save up for poor years. Trying to pin down an exact spending and investment strategy in retirement is a surprisingly slippery business.

  8. I also find it difficult to pin down an exact spending strategy in retirement.

    One idea which has some appeal me is to breakdown my spending into minimum and discretionary spending. Minimum spending are things like food, shelter, clothing, etc. and discretionary spending are things like traveling.

    Say a 1/4 of my ideal retirement spending is in this descretionary category that I could do without. Others will no doubt have their own breakdown.

    So then what scheme will provide reasonable certainty for the 3/4 of minimum spending with reasonable but less certainty for the 1/4 of discretionary spending?

    One scheme would be to set my withdrawal rate at 3% of the initial retirement savings, and only spend on discretionary when the retirement savings grows beyond the 3% + inflation.

    Of course, there is still the problem of deciding what to do if the market tanks in the first years of retirement. I can't imagine cutting my minimum spending back too much.

    One way to eliminate this risk is with inflation indexed bonds to cover off the minimum spending, and then put the remainder in stocks for the discretionary spending.

    It's a tricky business.

  9. Hey are you a professional journalist? This article is very well written, as compared to most other blogs i saw today….
    anyhow thanks for the good read!

  10. Blitzer68: Using bonds for a safe base and stocks for the rest is an interesting way to think of it.

    Anonymous: Thanks for the kind words. No, I'm not a professional journalist. I just like to analyze things, and I like money.