Category Archives: FI modelling

The 4% rule and early “retirement”

It’s not all beaches and sunshine

When posed with the question: “how large does my investment portfolio need to be in order for me to be financially independent/retire?” people will often quote the 4% rule, which means that if your yearly expenditure is $60,000 then you’ll need $60,000/0.04 = $1.5m to be financially independent. It also implies that if you withdrew 4% of the starting value of your portfolio each year to produce passive income you will never run out of money. But where does this rule actually come from? Also, how relevant is it for people that retire in their 30s and 40s?

To answer this question, I’ll first need to discuss sequence and longevity risk.

Sequence risk is the danger that the timing of withdrawals from a retirement account will have a negative impact on the overall rate of return available to the investor (Kagan 2019). In other words, because portfolio returns fluctuate, withdrawing money during periods of weak or negative returns can result in a portfolio not being able to last the duration of one’s retirement.

Longevity risk refers to the chance that life expectancy exceeds assumptions, resulting in greater-than-anticipated cash flow requirements (Kagan 2020). Put simply, there is a risk that you’ll run out of money in your portfolio before you die.

Defining withdrawal rate
For the purposes of this post, withdrawal rate will be defined as the percentage of the initial value of a retirement portfolio that is withdrawn to satisfy an investor’s income requirements. This value, in dollars, is then increased with inflation over time. For example, if an investor has a $1m portfolio and is using a withdrawal rate of 4% then they will withdraw $40,000 in their first year of retirement. If inflation is 2% for that year then in the second year, they will withdraw $40,800, regardless of the value of their portfolio in the second year. This means that the actual percentage of their portfolio that they are drawing down changes as their portfolio value changes.

Origins of the 4% rule
In 1994 William P Bengen published an article entitled “Determining Withdrawal Rates Using Historical Data” in the Journal of Financial Planning which looked at the longevity of portfolios constructed using varying amounts of stocks and bonds under withdrawal rates between 1% and 8%. He repeated this simulation using historical data for retirement portfolios commencing between 1926 to 1976.

One of the key findings is that for portfolios consisting of 50% stocks and 50% bonds; using a 4% withdrawal rate, the majority of portfolios lasted at least 50 years. This can be observed in the following figure taken from the article.

Ref: Figure 1(a) Determining Withdrawal Rates Using Historical Data (Bengen, W, 1994, pp173)

In addition to this, the minimum number of years that a simulated portfolio consisting of 50% stocks and 50% bonds lasted at a 4% withdrawal rate was 35 years. The following figure shows the minimum number of years that a simulated portfolio lasted under withdrawal rates ranging from 1% to 8%.

Ref: Figure 2 Determining Withdrawal Rates Using Historical Data (Bengen, W, 1994, pp 175)

For portfolios that consisted of 75% and 25% bonds the simulation actually resulted in a larger proportion of portfolios lasting the 50-year period, as seen in the following graph.

Ref: Figure 3(a) Determining Withdrawal Rates Using Historical Data (Bengen, W, 1994, pp 175)

In addition to having a higher chance of lasting 50 years; at the 20-year mark, the average value of a 75/25 stock/bond portfolio was higher than that of a 50/50 stock/bond portfolio, which bodes well for those with the objective passing their portfolios onto their children.

A second study conducted by Cooley, Hubbard and Walz from Trinity University draws parallels with the findings presented by William P Bengen. This study, entitled “Retirement Savings: Choosing a Withdrawal Rate That is Sustainable” looks at 30-year periods between 1926 to 1995 for withdrawal rates between 3% and 12%. It does this by using portfolios with various weightings of stocks and bonds.

The study draws conclusions that using withdrawal rates of up to 5%, at least 70% of the portfolios will last 30 years. Below are some the reproduced excerpts from the article that show the percentages of inflation adjusted portfolios that last 25 or 30 years given particular stock/bond weightings.

75% stocks/25% Bonds

Withdrawal rates3%4%5%
25 years100%100%85%
30 years100%100%86%
Ref: Excerpt from Table 2, Retirement Savings: Choosing a Withdrawal Rate That is Sustainable (Cooley, P, Hubbard, C, and Walz, D, 1998)


50% stocks/50% Bonds

Withdrawal rates3%4%5%
25 years100%100%79%
30 years100%95%70%
Ref: Excerpt from Table 2, Retirement Savings: Choosing a Withdrawal Rate That is Sustainable (Cooley, P, Hubbard, C, and Walz, D, 1998)


In both articles, when using a 4% withdrawal rate, a 75/25 stock/bond portfolio appeared to be more sustainable when compared with a 50/50 stock bond asset allocation.

To summarize, the 4% rule is grounded in rigorous studies that use data going as far back as 1926.

Additional considerations for early “retirees”

Retirement is a complex problem, where unpredictable events, and often changing circumstances influence portfolio requirements over time. “Retiring” early increases both sequence and longevity risk as longer time frames can be even less predictable than the already uncertain immediate future.

With that being said, now that we’ve looked at some evidence that supports the 4% rule; let me present some considerations when setting your own accumulation targets.

Transaction costs, fees and taxation
Both studies do not take into account transactions costs for performing activities such as rebalancing, in my previous article I highlighted this as an important consideration when looking to adjust your asset allocation. Taxation is also particularly important, especially if your cost of living is on the higher end. The table below shows the required pre-tax income that is needed to produce various levels of post-tax income in Australia for single person household.

Total portfolio distributions$30,000$50,000$70,000
Marginal tax bracket19%32.5%32.5%
Tax paid (inc. medicare levy 2%)$2,842$7,717$14,617
Take home$27,158$42,283$55,383
Ref: Resident tax rates 2020-21 (ATO, 2020)


For example, if your cost of living is around $42,000 then the 4% rule would imply that you require a $42,000/0.04 = $1.05m portfolio; however, when taxes on distributions are taken into account you will likely require a $50,000/0.04 = $1.25m portfolio. How you intend to create the income stream, may it be through the sale of portfolio units and/or taking distributions, will create different levels of taxation, all of which need to be taken in to consideration.

Simulations are based on past data
While studying the past helps us learn about how to deal with similar situations in the future there are no guarantees that history will repeat itself. Sticking religiously to the data means that we expect to see the similar occurrences in market movements, inflation and interest rates as we have in the past.

Currently we are experiencing some unique market conditions; interest rates and bond yields are currently at their lowest levels ever – we can’t be certain of what effect this will have on stock market or bond performance going forward. We could be heading into a period of low growth or the stock market may continue to grow at 8% p.a., it’s impossible to tell.

Early retirees have longer retirements
Freedom, it’s awesome, but of course it is not without risk. Typically, a 45-year-old Australian male today in 2018 is expected to live until 82.4 years; with females expected to live until 85.9 (AIHW 2020). This means that for a couple that retires at 45 years old their portfolio will likely need to last over 40 years. Given that the Trinity Study University Study only looked at periods up to 30 years, this represents a significant level of longevity risk when using a 4% withdrawal rate.

Expenses are likely to change
Both studies model for withdrawing a set amount, adjusted for inflation each year, however, in the real world, it is almost certain that your expenses will change year-to-year at a different rate to inflation. The good news is that it may not grow as quickly as inflation; however, if you are hit with a large unexpected medical bill then you will have no choice but to increase your withdrawal rate. As the studies have shown, withdrawing more than 4% of your 1st year portfolio value decreases it’s expected longevity by increasing sequence risk.

Asset allocations differ
Both studies use portfolios consisting of only stocks and bonds; it’s highly likely that your portfolio will have a different asset allocation. For instance, our portfolio contains REIT’s. This means that the risk profile and growth rates will be different to those presented in the studies.

What does this mean for us?

While some people will claim that the average person needs a giant $10m portfolio to retire because we need to protect ourselves from an endless list of negative scenarios, I personally think the 4% rule is a solid guideline for setting financial independence portfolio targets because of the rigorous analysis that supports it. However, this is based on the following conditions:

  • Targets are set based on fairly realistic budgetary spending patterns
  • Consideration is put towards how fees and taxation will affect your income stream during “retirement”
  • Target portfolios have a similar risk profile to those used in the studies
  • Early “retirees” are open to earning an income.

When setting targets, if you are very risk adverse then using a 3% withdrawal rate provides an even higher level of certainty than the 4% rule. This is supported by Bengen’s study where all simulated portfolios with a 50/50 stock/bond split lasted 50 years when using a 3% withdrawal rate.

Controlling for retirement challenges
Some of the considerations highlighted above represent risks to portfolio retirement planning, while they cannot be completely eliminated, they can be significantly reduced by: earning an income; controlling your expenses and ensuring that you are adequately insured. When combining these 3 initiatives you can greatly reduce sequence risk by decreasing the required the amount of portfolio distributions required; controlling day-to-day expenses and reducing the severity of unexpected expenses.

I want to finish here by saying that even though the 4% rule does not guarantee that your household’s expenses will be passively funded indefinitely the act of building a portfolio of such size is incredibly rewarding. The work required to do this builds a level of resilience that can help you weather more financial hardship than the average person that puts no thought towards their financial security and habitually wastes money. So, even if an early retiree finds themselves in situation where their portfolio can no longer provide them with enough income, the skills they have developed through planning, controlling expenses, increasing income, financial research, persistence, and hard work will put them in a far better position to deal with arising financial challenges.

Engineer your freedom

References

Kagan, J, 2019, Sequence Risk, Investopedia, viewed 4/11/2020, <https://www.investopedia.com/terms/s/sequence-risk.asp>

Kagan, J, 2020, Longevity Risk, Investopedia, viewed 4/11/2020, <https://www.investopedia.com/terms/l/longevityrisk.asp>

Bengen, W, 1994, Determining Withdrawal Rates Using Historical Data, Journal of Financial Planning pp 172 – 180.

Cooley, P, Hubbard, C, and Walz, D, 1998, Retirement Savings: Choosing a Withdrawal Rate That is Sustainable, Bierwirth Vol 10(1)

AIHW, 2020, Deaths in Australia, Australian Institute of Health and Welfare, viewed 4/11/2020, <https://www.aihw.gov.au/reports/life-expectancy-death/deaths-in-australia/contents/life-expectancy>

ATO, 2020, Individual income tax rates, ATO, viewed 5/11/2020, <https://www.ato.gov.au/rates/individual-income-tax-rates/>