Portfolio Stress Test

Portfolio Stress Testing

Retirement Plan Stress Testing

A Portfolio Stress Test—or a Retirement Plan Stress Test—is an important consideration when noodling the resiliency of your present assets in the light of future potential risks.

Future risks include known risks—market crashes, for one—and potential risks such as inflation and longevity. In addition, we live in an intermittently democratic society and legislative risk (changes in taxes or social security) is inevitable. And what about spending shocks such as health care and long-term care costs?

Stress testing current asset allocation against changes in market value or inflation is important. Beyond just your portfolio (as reflected by your asset allocation), you have a comprehensive retirement plan which integrates other sources of income such as social security and pensions. Retirement plan stress testing can help alleviate some of the anxiety that results from wondering if your money will run out in retirement.

Will you have enough? Let’s stress test your retirement plan and find out!

What is a Portfolio Stress Test?

Stress testing is a technique that models a portfolio’s response to various hypothetical scenarios. Using either past results with Monte Carlo odds or straight historical returns, you can see how various asset allocations perform given certain assumptions. Stress testing a portfolio looks at diversification.

Stress testing a retirement plan is more complicated, as there are various additional inputs such as social security, real estate, pensions, etc., that also have built in assumptions. Retirement plan stress testing is defined as calculating the odds of success of the entire plan given certain possible futures.

Certain possible futures imply different future risks.

What Retirement Risks should be Considered?

There are many retirement risks.

If we look at Dave’s Top 10 Retirement Risks, there are some risks that are easier to ponder than others. For instance, if you have inadvertent spending shocks or a local real estate crash, everything about your retirement plan can change. These, along with social risks such as cognitive decline and elder financial abuse, are difficult to plan for.

Other risks, such as sequence of return risk and market crashes, occur with regularity (but not predictability) and must be accounted for.

If you plan to retire in 2020, there are at least 18 retirement risks to think about. Which of these can we stress test and how difficult is it to do so?


Stress Testing for Specific Retirement Risks


Risks and considerations for retirement plan stress testing

Figure 1 (Risks and considerations for retirement plan stress testing)

Above, you can see different risks, how easy they are to simulate with stress testing, and suggested mechanisms for stress testing.

Those that are easy to model have variables simple to modulate on financial planning software.

Moderately difficult simulations include those with not only variable time frames, but also variable affects. So, for instance it is difficult to plan for unknown costs at a time uncertain.

Difficult-to-plan for risks include public policy risk as there is no way to predict the future. In addition, there are social risks such as frailty and financial elder abuse risk which take multimodal planning to address. Finally, Long-Term Care risk is perhaps the most intractable future risk present today for a segment of retirees with inadequate resources to self-pay but who don’t desire to spend down to Medicaid.

Let’s look at a retirement plan on professional planning software and see what we can learn about portfolio stress testing.

Retirement Plan Stress Testing Software

Professional planning software can be used to stress test a retirement plan.


Example of a retirement plan stress test

Figure 2 (Example of a retirement plan stress test)

Above, you can see a stress test based upon Monte Carlo probability.

At baseline, this plan has about an 80% chance of success. You can see it is most susceptible to inflation and taxes, and less susceptible to market crashes, social security changes, longevity, and increased health care costs.


Portfolio Stress Testing Severity

For this software portfolio stress test, you are able to increase or decrease the amount of stress you place on the portfolio.


stress testing stress levels

Figure 3 (Portfolio stress testing stress levels)

As you can see above, you can modulate the percentage of equity crash, tax expenses, social security income, and health care costs. In addition, you can change the severity of inflation from the given assumption and how much longer you and your spouse will live.

What happens if we double the stress to the portfolio?

Double Portfolio Stress Test

Double Stress Test

Figure 4 (double stress test)

Above, we have doubled all the stress test stresses from figure 3. As you can see, the baseline is unchanged.

Doubling the stress levels helps uncover inflation risk. Taxes and sequence of return risk (via market crash) are also highlighted as issues of concern.


How to Stress Test your Portfolio

DIY will want to know how to stress test their portfolio. I think Merriman has an excellent resource for this. You can use his fine-tuning table to see what different asset allocations do historically. Plan for a bit of a stress test of your own when you look at them. I like to simplify them a bit.

Portfolio Stress Test without Software

Figure 5 (Portfolio Stress Test without Software)

For your selected portfolio’s asset allocation, you can see the annualized return and standard deviation. Also, since 1970, see the worst 3-, 6-, and 12-month returns, as well as the worst 3- and 5-year annualized returns.

Using this fine-tuning table, a DIY investor can see how well diversified his or her asset allocation is. This is a stress test for your portfolio you can do without software.


How to Stress Test your Retirement Plan

How to Stress Test your Retirement Plan

Figure 6 (How to Stress Test your Retirement Plan)

For DIY investors, good luck with stress testing your retirement plan. Consider using an Advice-Only Financial Planner for a one-time financial plan.



Stress testing is important in medicine and for the banks. Why not stress test your portfolio? Stress testing your portfolio can help determine if your asset allocation is well diversified for your risk tolerance.

Or, when you are closer to retirement, consider stress testing your retirement plan. Retirement plan stress testing is important to help anticipate any possible future gaps in the plan, and to ensure your plan will hold water no matter the risk.

Posted in Retirement Income Planning and tagged .


  1. Great topic, but a bit frustrating that you didn’t finish the article with at least a few suggestions on (software, calculators, etc.) options for actually stress testing a variety of retirement scenarios that you mention. I really like your work, so please take this as constructive criticism. I would venture a good number of your readers are DIY investors, so just finishing by saying “DIYs…good luck…” was a slightly disappointing end to the article. Maybe a follow up on exactly how to do a variety of DIY stress tests would possibly supplement this piece? Just a suggestion…Again, I do appreciate your articles, and this is a great topic! I jjust thought this particular article ended before it was fully finished. Would love to read a follow up!

    • ThomH,

      Thanks for the feedback. I was a little flippant. Stress testing is pretty hard for DIY. I’ll try to look into resources on-line and see if I can find some better advice than I gave above!

      • Thanks for responding. I reread my earlier comment, and I hope my comment didn’t come across as critical. I am a regular reader, so I greatly appreciate everything you post, and all of your efforts to provide outstanding information!

  2. The issue isn’t so much stress testing, it’s efficient planning and correct framing. Consider a Gaussian curve. In accumulation you aim for the mean return if you are a passive index investor, the center of the curve. In retirement however you worry about the risk, which lives in the tails not the center. The upper tail you can ignore sine that tail represents better than average so risk aversion is concentrated in the lower tail. What you want to pay attention to is the “fatness” of the lower tail. A skinny lower tail means the likelihood of running out of money is small. A fat lower tail means the likelihood is much higher.

    You can stress test this by using the monte carlo calc in portfolio vizualizer. You are free to change WR portfolio size, asset allocation, longevity, inflation. withdrawal model, and a form of SORR stress testing where the worst of the sequences are put in the front of the line eg you can put the worst 5 years first in a 30 year portfolio. The calculator reports out in % success and when the first failure occurs. It also displays a Gaussian result for various confidence levels of survival and safe WR and perpetual WR plus a lot of other stats. It’s about as good as a DIY is going to do. You can use the efficient frontier calc on this site to first analyze the most efficient AA. Between the EF calc and the MC calc you can forecast as good as anybody, except this calc doesn’t take into account taxes and diversification across tax accounts.

    I’ve had no trouble creating a completely stress tested and optimized portfolio using this software and spread sheets of my own design to iterate the most bang for the least risk. It does take time and commitment, but it’s easy enough to do. The advantage is you have only one portfolio to analyze so a brute force method is perfectly adequate to reach a conclusion for a single portfolio. If I was doing dozens I’d need more automation and proof of performance to validate the results. One can be done by hand.


    Other calcs I use are




    and EXCEL

    • Gasem, Thank you for taking the time to respond and for the links. Great points. I’ve used similar calculators, but I do like some of the flexible features in the Portfolio Visualizer that you provided. (I’ll be running simulations for days now!) I’ve been running similar simulations (via Excel and other Monte Carlo simulators using pure brute force) for years, double checking my overall plan periodically. My portfolio (like many) has complications of significant real estate and various investment categories beyond typical stocks and bonds, as well as the typical tax complications of any portfolio, so was I guess I was just hoping to find more “uniquely useful” tools, that I hadn’t yet come across. But it sounds like brute force is still the best option at this time! …oh well, there goes the rest of my day! 😉

      Thanks again David and Gasem. You guys are doing a huge service for the FIRE community! Keep up the great work!

Comments are closed.