How much money should I withdraw annually from my portfolio when I retire?
I get that question a lot from friends and family. (Occupational hazard.) It’s also one of the most hotly debated issues in financial planning. Why? First, it’s important; we all hope to live happily in retirement. Second, every person’s situation is unique, so there’s no standard set of spending assumptions for retirement planning. Third, market returns may be mean-reverting over long time periods, but a person’s retirement happens over a specific time period, parts of which may deviate significantly from longer-term average returns that are used to forecast future asset values.
Let’s start with why the question is so critical. Ideally, you’ve been saving for four to five decades to build your nest egg. Now that you’ve stopped working, you want to use that hard-earned money for daily expenses, health care and the things you wanted to do while you were working — like taking a month-long African safari. But you also want to make sure your money lasts until you or your spouse dies, whichever comes later. Often, you want it to last even longer: Many people hope to pass along some of their assets to their children, grandchildren and other loved ones.
Now, if we all had the same assets, spent roughly the same amount each year and died at the same age, figuring out how to budget for retirement would be a relatively simple process. Of course, we don’t. So the retirement planning equation involves many variables, such as: At what age do you want to retire? Do you have a pension or expect to receive Social Security? What are your fixed expenses and how much will they increase with inflation? Do you have a family history of longevity? Do you want to pass money down to subsequent generations?
This is a just a sample of typical retirement planning questions, but it makes the point: We all have different starting assumptions, needs and desires. Yet, many people of different means often get the same answer when they ask how much money to withdraw annually from a retirement portfolio: Usethe 4% rule.
Originally articulated in a 1998 paperbythreeprofessorsatTrinityCollege in Texas (a paper now known as the TrinityStudy), the rule recommends that you withdraw 4% of your retirement savings in your first year of retirement. Every following year, withdraw the same amount plus an adjustment for inflation. Historically, retirees who follow that schedule have had a high chance that their portfolio will last at least 30 years. In the original study, the success rate varied slightly depending on the ratio of stocks to bonds in a portfolio, with a 50-50 stock to bond ratio producing a 95% success rate. (Subsequent studies have examined how to add a variable component to the formula based on previous year portfolio returns with the goal of increasing the success rate to 100%.)
The 4% rule is a simple one and easy to remember, which I think is one reason why it’s become so popular in my business — it’s a little like a doctor telling a patient to take two aspirin and call in the morning. Another is its high estimated success rates; again, like doctors, financial planners first want to do no harm.
But in the years since the rule was introduced, skeptics have raised a number of objections to it. Some say that it’s too conservative and that it will cause those who follow it to live less well than they would like to, particularly in their early retirement years when people are more likely to travel and spend more. Others say it’s too aggressive: What happens if we’re in an extended period of stock market stagnation or decline — like, say, the last decade? Shouldn’t you then be withdrawing less than 4% when 10-year government bonds are paying less than 2%? Put simply, should your withdrawal rate remain constant when your returns don’t?
Nobody wants to be a member of the unlucky 5% of retirees who follow the 4% rule but still run out of money. In my next post I’ll look at some of the ways economists and financial planners have suggested amending the rule to prevent that from happening.