Safe Withdrawal Rate: Dual Life Expectancy Planning

In the prior article we reviewed our Safe Withdrawal Rate (SWR) for our scenario.  We are planning for a 4% SWR with an annual expense of $72,500.00 in today’s dollars adjusting for inflation as we retire.  This comes out to a Financial Independence savings goal, Our Number, of about $1.8 million dollars. 

Based on our research and simulation for 45 years of retirement that worked out to a 98.6% chance of success based on our assumptions made in the Monte Carlo Simulation tool.   Additional research also suggests the Monte Carlo simulation may overstate both the worst and best performances since the models do not address mean regression as seen in the real world.  Per the study by Derek Tharp, a 93.5% probability is equivalent to 100% in the real world1, so we may be even better off than the simulation shows.

So far this all looks promising.  We can make different assumptions on inflation and market performance and I am sure there are other opinions on how we chose our basis for the model.  We have a plan with a reasonable level of risk we can manage.  As we discussed in the prior article, we are not going to set this forever and not make any adjustments we will keep an eye on the markets and make decisions on when we pull the trigger and once retired will have control over our withdrawal rates and expenses.  We do not have extravagant budgets but have some cushion in our plans.

Sounds like we are all set, but Mrs. FoF asked a good question – “What happens if I out live you?” 

How do We Estimate our Combined Retirement Length?

Mrs. FoF is several years younger than myself.   She also has good genes on her side, the females in her family have had long active lives. So how do we determine a good number for our combined life expectancy?    In comes, Google and some quick searching show Vanguard has developed an online tool, Plan for a long retirement2, just to help answer this question.

The Vanguard tool allows the user to input the current ages for two people and predict the probability they will last to a certain age.  The mortality data is from the Society of Actuaries Retirement Participant 2000 Study3

One decision we had to make is what survival rate would we use as our planning cut off, after some discussions we came to 5% of less for Mrs. FoF.     This is another decision you will need to make with your spouse or significant other and what works for us may not be the best for you. This is sometimes a little difficult to discuss, talking about when you plan to die but to ensure success for our Financial Independence journey it’s a conversation we need to have. 

Predicted Dual Life Expectancy

Below is our result, to get below 5% we need to plan for a 55-year retirement.    Looks like I have 0% chance to living to above 100 but Mrs. FoF has a 4% chance!

Dual Life Expectancy (+95% Probability)

One thing to check is the <5% probability for Mr. FoF and to see if we can adjust the income needs once Mr. FoF life expectancy is over.  Two factors for this is as we age we tend to spend less money, more on this in the future and one person usually has lower income needs than two.

Mr. FoF Life Expectancy

We now have two data points.  We need to plan for at least 55 years of retirement and after retirement year 47, we can consider reducing our income needs to support only one person.  Note: 47 years is even longer than my initial estimate of 45, so it was good we looked. 

So now back to our Monte Carlo Simulation to test out this new situation!  For more details on the Monte Carlo Simulator see our prior article 

The 4% Safe Withdrawal Rate – Validation with Monte Carlo Simulations for Our Scenario

Safe Withdrawal Rate Test Case 1:   55 Years of Retirement

Note: A  bug was found in the Simulation Tool where the average investment performance was fixed to 9.5% with a standard deviation of 8.9% for both pre and post retirement simulations.  This was roughly about 50/50 stock/bond performance using 50 years of historic data.  Results are still valid, just note the portfolio ratio is not as shown below.

First thing we did was just re-run the Monte Carlo Simulation with all the same assumptions but with a 55-year retirement window.  This gives us a baseline on how well the 4% SWR holds up for this plan. 

1,000 Monte Carlo Simulation of 55 Years of Retirement

Interestingly our probability of success only dropped from 98.6% to only 98.2%, well within our comfort zone.   I re-ran this for 10,000 cycles just to make sure we didn’t get a lucky simulation.   The result below we similar with a 98.0% success rate.

10,000 Monte Carlo Simulation of 55 Years of Retirement

Safe Withdrawal Rate Test Case 2:  55 Years of Retirement, Reduced Income Need for Surviving Spouse

We wanted to investigate reducing the income need by 40% once only one spouse needed income.  Based on the first test case we did not expect much change in the performance since the success rate was very close to the 45-year plan.

One benefit of using Excel is adjusting the model is straight forward.  Another reason I like having this in Excel vs. depending on an online tool.  We added an options page to allow some tweaking of the model assumptions.  The first option is added the older spouse expected retirement length and the % income need reduction for the remaining spouse.    We also tweaked the input page to change from life expectancy age to overall retirement length to account for the longest life expectancy.

Added Dual Life Expectancy Option to Monte Carlo Simulator

Compared to the prior test case, we did not expect to see too much difference.  The model for 45 years and 55 years did not vary too much and  the probability did not really change.  This was another run of 10,000 cycles and produced a 98.3% success probability.

Dual Life Expectancy Income Adjusted Simulation

Do We Need to Adjust Our Safe Withdrawal Rate?

Based on extending our retirement window for our scenario, the 4% rule still seems to hold up.  The simulations show at least a 98% chance for success for our 55-year plan.  This is not too surprising since other research suggeststhe 4% SWR should last almost indefinitely.

In a recent AMA Reddit4 post by Mr. Bengen5, he comments on longer retirement time horizons based on his latest studies.  Here is an excerpt below:

” As your “time horizon” increases beyond 30 years, as you might expect, the safe withdrawal rate decreases. For example, for 35 years, I calculated 4.3%; for 40 years, 4.2%; and for 45 years, 4.1%. … If you plan to live forever, 4% should do it.”

Sequence Risk and Next Steps

My guess based on these simulations is once we get past the sequence risks of some of the worst-case cycles.  Once our money survives the first 10 or so years, the chance we run out of money is very low regardless of retirement length.

We will discuss and review sequence risk in more detail in a future article since it is something I am very concerned. We are entering a phase of the market where sequence risk has a higher likelihood of being a concern.

Sequence Risk – Will It Effect Your Retirement Plans?

Once thing we found in doing our own research is as many people who say a given Safe Withdrawal Rate is good, some say it may be too conservative and others say it may be too risky.   Our research suggests for our situation we would are comfortable with a 4% SWR. Our assumptions and numbers only apply to our situation and you will need to review your own needs, assumptions and risk tolerance for your own journey towards Financial Independence. 

Right now, Our Number looks good based on our risk tolerance and we have flexibility in our timing and plans during retirement. 

If you subscribe to the blog, we are planning to send out a limited number of copies of the Monte Carlo simulation spreadsheet for Beta testing.  If interested in Beta testing the spreadsheet, just subscribe below.  We will send out an invite email for the Beta testing to get some feedback on the tool.

Sign Up Today!

Get the latest version of the Feeding Our Fire Retirement Simulator!

References and Links

  1. Does Monte Carlo Analysis Actually Overstate Tail Risk In Retirement Projections? , Derek Tharp, 2017 
  2. Plan for a Long Retirement, Vanguard 
  3. The RP–2000 Mortality Tables, Society of Actuaries 
  4. I’m Bill Bengen, and I first proposed the 4% safe withdrawal rate in 1994. Ask me anything! Reddit AMA Post from William Bengen, 2017
  5. Determining Withdrawal Rates Using Historical Data William Bengen, 1994,  

Other Articles in the SWR Series

If you subscribe to the blog, we are planning to send out a limited number of copies of the Monte Carlo simulation spreadsheet for Beta testing.  If interested in Beta testing the spreadsheet, just subscribe below.  We will send out an invite email for the Beta testing to get some feedback on the tool.

You Might Also Like