A Critical Look at the "Multiply by 25" Rule to Early Retirement
An actuarial perspective on the determination of your retirement number.
The Multiply by 25 rule is derived from the Trinity Study, which concludes that you can withdraw 4% of your investment portfolio each year without incurring a substantial risk of running out of money. As a result, you could (theoretically) retire if you reach an investment portfolio of 25 times your annual expenses.
As an actuary, I have two immediate concerns with regard to the Multiply by 25 rule. First, the underlying assumptions for this rule are based on 3% inflation and an annual return of 7% from your investment portfolio, which results in a rate of return of 4% (above inflation). Given the current economic environment, I wonder whether a consistent rate of return of 4% (above inflation) is achievable without taking unnecessary risk. However, for the purpose of this post, let's assume an inflation of 3% and an annual return of 7%.
My second concern is regarding the 25x multiplier, this multiplier is determined based on a growing perpetuity formula, which is elaborated below.
From an actuarial standpoint, determining your retirement number based on the 25x multiplier leads to an overestimation.
If the assumptions of this multiplier hold true, the growing perpetuity formula implies that you would never run out of money, which is certainly desirable. However, you don't need to save up that much for retirement, as the probability of someone living for eternity is 0. As a result, the biggest shortcoming of the Multiply by 25 rule, is that it does not consider an important factor, which is the expected lifespan of an individual.
Imagine that the expected lifespan is 75 years and there are two individuals that want to determine their retirement number, a 30 and 50-year-old. Based on the Multiply by 25 rule, if their annual expenses are equal, then their retirement number would also be the same. For example, if the annual expenses of both individuals are $30,000, then their retirement number is $750,000 (25 x $30,000).
This should not be the case as the 30-year-old has an expected remaining lifespan greater than the 50-year-old, which is respectively 45 years compared to 25. As a result, the retirement number of the 30-year-old should be greater than the retirement number of the 50-year-old.
Actuarial retirement multiplier
The determination of the actuarial retirement multiplier is based on discounted probability-weighted cash flows. In layman's terms, this means determining the present value of a stream of payments considering all future uncertainties.
For simplicity, I will only consider the expected lifespan of one individual, by adjusting the future cash flows with the probability that he/she stays alive. These probabilities are derived from the most recent life table provided by Social Security Association (SSA).
The multipliers presented above are considered as the actuarial equivalent of the Multiply by 25 rule. The only difference is that these multipliers account for the expected lifespan of an individual as opposed to the Multiply by 25 rule. To illustrate, the retirement number of a 35-year-old male and female is determined by multiplying their annual expenses with, respectively, the actuarial multipliers 19.84 and 20.75.
Comparing these actuarial multipliers with the Multiply by 25 rule reveals some interesting observations:
- First, none of the actuarial multipliers are greater than the 25x multiplier, which is expected as the Multiply by 25 rule is based on the growing perpetuity formula and does not consider the expected life span of an individual.
- Another observation is that the actuarial retirement multipliers for females are in general greater than the actuarial retirement multipliers for males, which is also consistent with the fact that women generally live longer than men.
- Lastly and most importantly, there's an inverse relationship between age and the actuarial multipliers. As an individual becomes older, their remaining (expected) lifespan decreases and therefore their early retirement number should also decrease. This is expressed in the decreasing actuarial multipliers per age.
Conclusion
Let me preface this by saying that long-term forecasting is extremely challenging. Assumptions need to be carefully selected on a best estimate basis, as a small change in these assumptions can have a significant impact on the outcome.
In conclusion, despite my reservations regarding a consistent and long-term rate of return of 4% (above inflation), I've shown that applying the Multiply by 25 rule not only overestimates the retirement number that an individual would need, but also highlighted the importance of considering the expected lifespan of an individual in determining the actuarial retirement multiplier and subsequently their retirement number.
The financial opinion expressed in this article are from personal research and experience of the writer. This article is for informational purposes only. It should not be considered Financial or Legal Advice.
