Data and Assumptions for Wendy
Wendy is 22 years old and would like to retire at age 62. She wants to contribute $3,600 per annum increasing by 3.5% per year (to her employer’s 401(k) plan). We will assume that she earns $72,000 per annum (so her contribution is 5% of her pay) and expects her pay to increase by 3.5% per annum. We also assume that Wendy’s employer will match Wendy’s contributions $.50 on the dollar up to 6% of her pay.
We will also assume:
- 5% annual inflation (1% less than Wendy’s assumed annual pay increase assumption)
- No part-time employment after retirement
- No other sources of income
- No long-term care costs
- No increased medical care costs after retirement and prior to Medicare eligibility
- No desired estate
- No unexpected costs or non-recurring expenses
- Estimated Price of an inflation-adjusted life annuity used at retirement to support Wendy’s recurring spending budget at retirement (2% real rate of return and 25% probability of survival for a healthy non-smoking female). This is consistent with our default investment return recommendation and coincidentally, is a practice endorsed by Mr. Fox in his article.
- Projected Social Security benefits payable under current law based on future dollar results from the Social Security Quick Calculator program based on current pay of $72,000 and 3.5% per annum future pay increases.
Mr. Fox discusses several goals that might be considered by Wendy. We agree with Mr. Fox that accumulating a specified amount of assets is unlikely to be consistent with Wendy’s underlying retirement objectives. In our opinion, a much more robust goal is to target after-retirement spending to be about the same as before-retirement spending when adjusted for expected reduced taxes and work-related expenses. We have used 85% of pre-retirement spending as the target for the tables below. If Wendy has significant travel plans or other expenses after retirement, she may want to target a somewhat higher after-retirement spending replacement percentage.
Pre-Retirement Investment Return
Mr. Fox assumes an 8% per annum average investment return with a 12% standard deviation around that average to generate his Monte Carlo simulations. This is an important assumption and is essentially a 6.5% real rate of return assumption under the assumptions above. Many would (and have) argued that while this real rate of return may have been achieved historically, it is probably overly optimistic in the current investment environment.
Retirement at 62. Contribution rates required under different assumed pre-retirement investment returns
Based on the assumptions and data above, let’s look at the contribution levels/savings rates (measured as a percentage of Wendy’s pay and rounded to the nearest 0.5%, not including matching contributions) to achieve a post-retirement spending budget of 85% of pre-retirement spending for various assumed real pre-retirement investment returns. This table assumes Wendy wants to retire at age 62 and will commence her Social Security benefit at age 70.
| Real Rate of Return per annum | 0% | 2% | 4% | 6% | 8% |
| Savings Rate Required | 20.0% | 15.0% | 10.0% | 6.0% | 4.0% |
For those of you who would like to crunch numbers, here is the screen shot from our ABC for single pre-retirees used to develop the 10% contribution requirement for Wendy under the 4% real (6.5% pre-retirement investment return minus 2.5% inflation) scenario.
 |
| (click to enlarge) |
Pre-retirement investment return of 4% real. Contribution rates required under different ages of retirement
Wendy now wants to see how different assumed retirement ages could affect her annual savings rate. Based on the assumptions and data above, let’s look at Wendy’s annual contribution rates to hit the same target 85% spending replacement target at various retirement ages assuming a 4% per year real rate of return.
| Retirement Age | 55 | 62 | 65 | 67 | 70 |
| Savings Rate Required | 17.5% | 10.0% | 7.5% | 6.0% | 4.0% |
Communicating risk and enabling clients to make better financial decisions
Stochastic (Monte Carlo) modeling is not the only way to communicate risk of plan failure. And if the Stochastic model uses optimistic assumptions about the future, it can be a poor tool for this purpose. Scenario or stress testing of deterministic models or developing spending budgets based on low-risk investments can be just as effective in quantifying risk and communicating useful information for informed decision making.
Most important component of a retirement plan
Yes, the ABC workbook models we make available on our website use deterministic assumptions. Our models could, of course, use stochastic assumptions for different types of investments, but we are comfortable recommending relatively low-risk deterministic assumptions as risk-adjusted results of using more complicated stochastic assumptions should be similar. Using relatively simple deterministic assumptions makes our ABC models much more accessible to individuals who want to do their own budgeting and modeling.
If assumptions used in our models (or stochastic models for that matter) are not accurate, future spending budgets/required savings rates will be automatically adjusted (the same, up or down). This automatic adjustment process tells Wendy, for example, how she is doing from year to year. For example, if year to year required savings rates are increasing, this is an indication that Wendy’s assumptions about the future are too optimistic and may need to be made more conservative. This automatic adjustment process takes place either by evolution (through higher required savings rates) or by revolution (through assumption changes).
In our opinion, the most important component of a person’s retirement plan is the actions they will take when actual experience inevitably deviates from assumed experience. The Actuarial Approach sets forth a proven process for this necessary adjustment.
Wendy decides
Our website is all about using basic actuarial principles to help people make better financial decisions, so inquiring minds want to know: What did Wendy decide? Based on her research, Wendy believes a 4% real rate of return is probably more realistic in today’s economic environment, but she doesn’t think she can afford to contribute 10% of her pay at this time. She understands that it is important to maximize her employer’s 401(k) contribution by contributing at least 6% of her pay and she still wants to target age 62 for retirement. Therefore, she decides initially to assume the 6% real rate of return pre-retirement assumption (8.5% investment return/2.5% inflation). She understands that her calculated savings rate may change as actual experience emerges. By scenario testing various possible decreases in her projected Social Security benefit, different future employment patterns, increases in expenses resulting from long-term care needs etc., she is well aware that she may have to increase her savings rate in the future in order to achieve her retirement goals (which will undoubtedly change over time).
Summary
Instead of finding Mr. Fox’s Monte Carlo analysis to be a powerful tool for understanding, we found it confusing. For example, it didn’t seem to develop a reasonable spending goal and it didn’t seem to indicate how much the client should contribute in order to achieve her goals based on reasonable assumptions. Finally, it didn’t seem to indicate what her plan would require when actual experience inevitably differed from assumed experience. We suggest that you try our more accessible, straight-forward Actuarial Approach to help you make your retirement financial decisions instead.
