Observations: Assumptions for the "4% Withdrawal Rate" Retirement Savings Graphs

Retirement planning: retire safely & comfortably using 4% initial withdrawal rate

The Observations posts based upon the Trinity Study / "4% Withdrawal Rate" guidelines for retirement savings include a short version of the assumptions made. This post discusses the assumptions underlying those graphs in more detail. (See Related Materials below for a list of the posts.)

The Trinity Study (The "4% Withdrawal Rate" Approach to Determining Retirement Savings Needs)

The Trinity Study was a retirement study conducted by three finance professors from Trinity University. They assumed that, given today's lifespans, a safe retirement portfolio should be large enough to last for thirty years. Their methodology involved simulating the performance of a variety of stock/bond allocations using actual historical market performance data. Their research concluded that the most important factor in having retirement savings last that long was a realistic initial withdrawal rate. Further, they found that for a wide range of portfolios, a portfolio would have lasted 30 years more than 90% of the time if you started with a 4% initial withdrawal rate.

The portfolios were tested by withdrawing 4% of the portfolio in the initial year of retirement. Thus, the portfolio initially held 25 years worth of expected expenses. Withdrawals were increased by inflation each year after that to fund the hypothetical retiree's yearly expenses. They defined success as not running out of money before 30 years had elapsed.

The hypothetical portfolios ranged from 100% stocks and 0% bonds to 0% stocks and 100% bonds. Results were calculated based on actual market performance in the forty 30-year periods from 1926 to 1995. The portfolios that were at least 50% stocks were successful more than 90% of the time; in addition, they supported a 5% withdrawal rate more than 75% of the time.

While this period is broadly representative, there is no guarantee that future market returns will be as large, or that your post-retirement returns will match market returns. (Also, these were hypothetical portfolios; as far as I know, they did not include expenses.)

Their Study Assumes Retirees Receive Neither a Pension, nor Social Security!

Note that the Trinity study assumes the investment portfolio is the retiree's only source of income. However, where noted, my graphs include the impact of Social Security. On the other hand, because so few people are eligible for traditional pensions, my graphs do not account for their impact. If you have a pension that includes cost of living adjustments you can calculate the impact by using My SIMPLE Retirement Savings Calculator, and combining your Social Security income with your pension income.

Additional Assumptions for the Observations Retirement Savings Graphs

The graphs in the "what % should I save" and "how much should I have in savings" series assume that the goal is to save enough money by your retirement date to follow the 4% withdrawal approach. (See Related Materials below for links to the graphs.)

In order to calculate these benchmarks, I've assumed that you are starting your retirement savings plan "from scratch." That is, you currently have no significant retirement savings. (However, if you already have savings, see links below to the "How Much Should I Have In Retirement Savings?" series for a way to use these same benchmarks.) In addition, I assume that once you start, your salary will increase at the same rate as inflation, and you will contribute a constant percent of that salary each year until you retire (i.e., your contributions also increase with inflation).

My Social Security benefits were calculated based upon then current rules; these rules may changed since then, but typically the basic shape of the curve remains the same. Combining the salaries of more than one Social Security beneficiary will result in my underestimating the benefits that are due (under current rules); do each wage-earner separately. The "w/o Soc Sec" graphs don't have this issue. (See below re changing the assumptions.)

The graphs show results if you earn pre-retirement returns of 3, 5 and/or 7% -- after inflation, taxes and investment expenses. If inflation averages 3%, "5% after inflation" is equivalent to the 8% nominal return often assumed by financial planners. Historically, this would typically have required that at least 50-60% of your assets be in stocks. However, even so, there is no guarantee that you will earn your planned pre-retirement returns. No matter what returns you plan, I suggest you monitor your actual progress along the way and make adjustments as appropriate! (See more on "real" -- i.e., after-inflation -- vs nominal returns.)

Finally, I've assumed you will retire at age 65, and want about 75% of your current annual salary, adjusted for inflation, each year in retirement. (Don't forget, you'll no longer be saving a big chunk of your salary for retirement.)

What If I Want to Change Some of the Assumptions?

For an estimate more tailored to your specific circumstances, and to get a better feel for the impact of some of the variables, I encourage you to enter your own data into My SIMPLE Retirement Savings Calculator. Specifics of your situation, such as your expected pre-retirement investment returns, expected Social Security benefits, and expected spending level in retirement, could make a significant difference.

Related Materials:

Start Retirement With a 4% Withdrawal Rate A discussion of the 4% withdrawal concept, from Time Magazine. For a more detailed discussion, see Wikipedia.
How Long Will You Live? A look at one of the most vexing issues in retirement planning.

Real World Expenses Reduce Published Market Returns : on real world vs hypothetical returns

Easy-to-use graphs based on the Trinity Study/"4% withdrawal" approach.

My SIMPLE Retirement Savings Calculator interactive calculator using these same calculations.

For lists of other popular posts and an index of stock market posts, by subject area, see the sidebar to the left or the blog header at the top of the page.