Is Typical Retirement Advice Good Advice? – Testing Retirement Rules of Thumb (As seen in Canadian MoneySaver)
The “4% Rule” only works with at least 50% in stocks.
You want to retire soon. How should you setup your retirement income?
You talk with some friends, read about it on the internet, and talk with a financial advisor. Are you actually getting good advice?
When it comes to retirement income, most financial advisors rely on a few rules of thumb handed down from one generation of advisors to the next. The rules appear to be common sense and are usually accepted without question.
Do these rules of thumb actually work?
Before giving clients this advice, I tested them with 150 years’ history of stocks, bonds and inflation. I wanted to see if these rules were reliable for a typical 30-year retirement. (The average retirement age is 62. In 50% of couples that reach their 60s, one of them makes it to age 92.)
These 5 rules are the “conventional wisdom” – the advice typically given to seniors:
- “4% Rule”: You can safely withdraw 4% of your investments and increase it by inflation for the rest of your life. For example, $40,000 per year from a $1 million portfolio.
- “Age Rule”: Your age is the percentage of bonds you should have. For example, at age 70, you should have 70% in bonds and 30% in stocks.
- “Sequence of returns”: Invest conservatively because you can’t afford to take a loss. You can run out of money because of the “sequence of returns”. You can’t recover from investment losses early in your retirement.
- Don’t touch your principal. Try to live off the interest.
- Cash buffer: Keep cash equal to 2 years’ income to draw on when your investments are down.
The results: NONE of these rules of thumb are reliable, based on history.
Let’s look at each to understand this.
- “4% Rule”: Can you safely withdraw 4% of your investments plus inflation for the rest of your life?
Based on history, the “4% Rule” was safe for equity-focused investors, but not for most seniors.
In the results below, the blue line is the “4% Rule”, showing how often in the last 150 years a 4% withdrawal plus inflation provided a reliable income for 30 years.
The “4% Rule” only works with at least 50% in stocks.
The “4% Rule” worked only if you invest with a minimum of 50% in stocks. Even safer is 70-100% in stocks. It is best to avoid a success rate below 95% or 97%. They mean a 1 in 20 or 1 in 30 chance of running out of money during your retirement.
Most seniors invest more conservatively then this and the 4% Rule failed miserably for them.
A “3% Rule” has been reliable in history, but means you only get $30,000 per year plus inflation from a $1 million portfolio, instead of $40,000 per year.
These results are counter-intuitive. The more you invest in stocks, the safer your retirement income would have been in history.
To understand this, it is important to understand that stocks are risky short-term, but reliable long-term. Bonds are reliable short-term, but risky long-term. Why? Bonds get killed by inflation or rising interest rates. If either happens during your retirement, you can easily run out of money with bonds.
The chart below illustrates this clearly. It shows the standard deviation (measure of risk) of stocks, bonds and cash over various time periods in the last 200 years. Note that stocks are much riskier short-term, but actually lower risk for periods of time longer than 20 years.
Stocks are more reliable after inflation than bonds after 20 years.
Ed’s advice: Replace the “4% Rule” with “2.5% +.2% for every 10% in stocks Rule”. For example, with 10% in stocks, use a “2.7% Rule”. If you invest 70% or more in stocks, then the “4% Rule is safe.
- “Age Rule”: Your age is the percentage of bonds you should have. For example, at age 70, you should have 70% in bonds and 30% in stocks.
History shows that the “Age Rule” directly conflicts with the “4% Rule”. Based on the “Age Rule”, a typical retirement from ages 62 to 92 would average almost 80% in bonds. The “4% Rule” ran out of money 31% of the time with 80% in bonds.
Ed’s advice: The “Age Rule” works with a “3% Rule”. There is nothing wrong with investing conservatively with the Age Rule, but then reduce your retirement income to withdraw only 3% of your investments each year.
- “Sequence of returns”: Should you invest conservatively to avoid losses because you can’t recover? Will you run out of money if you have some investment losses early in your retirement?
The “sequence of returns” is not supported at all by history.
Can you be confident in the stock market? Over the long-term – yes (based on history). Short-term or medium-term – no.
The chart below shows actual history of the 4% Rule with 100% in equities. Each line is a 30-year retirement. Note that there was a market recovery within a few years of the majority of market declines, even though you continued with the same retirement income and increased it every year by inflation.
There were quite a few market crashes in the last 150 years, but only once would you have run out of money because of a market crash (retiring in 1929).
There are 118 retirements of 30 years on this graph. Investing 100% in stocks, you would have run out of money only 5 times – 4 because of very high inflation and only one because of a market crash.
Retirement Income Withdrawals in History – 100% Equities & 4% Withdrawal + Inflation
“4% Rule” with 100% equitties – 96% success in history.1
Because of the “sequence of returns”, the typical advice is to invest more conservatively with more bonds. This is not safer!
The chart below is the 4% Rule with 100% in bonds. There are 118 retirements of 30 years on this graph. Investing 100% in bonds meant you would have run out of money in 63 of them – more than half the time! If you retired almost any year between 1890 and 1980, you would have run out of money with 100% in bonds. Note how many of these lines drop below $0:
Retirement Income Withdrawals in History – 100% Bonds & 4% Withdrawal + Inflation
“4% Rule” with 100% bonds – Only 47% success in history.1
Ed’s advice: Invest within your risk tolerance. But do not think more bonds make your retirement safer. If you invest more in bonds, reduce your retirement to a reliable level of 2.5-3% withdrawal rate. Fixed income is lower income.
- Don’t touch your principal. Can you live off the interest?
This rule of thumb completely ignores inflation. In a typical 30-year retirement, the cost of living triples.
This means you effectively take a cut in your income every year. Over a 30-year retirement, your income has dropped to only 1/3 of your income in the first year.
When you hear seniors complaining about being on a “fixed income”, it is usually because they are trying to live off the interest.
Ed’s advice: Any reasonable retirement income plan needs to include consideration for inflation increases most years.
- Cash buffer: Is it safer to keep cash equal to 2 years’ income to draw on when your investments are down?
This seems to make some sense, because you can live off the cash whenever your investments are down, giving them time to recover. You can avoid selling when prices are low, right?
Sorry. History does not support the Cash Buffer at all. “No cash” has consistently been the safest:
Highest success rate is No Cash buffer. It is safer NOT to hold cash.
In fact, I have been unable to find even a single example where holding any amount of cash was safer than no cash. I found studies using global stocks and UK stocks with many different strategies of how to use the cash. No study I found had even a single example of a benefit of cash.
The drag on your returns from holding cash sometimes caused you to run out of money, but holding cash never protected you from running out of money.
In addition, if you held cash, you died with a significantly smaller estate to pass on to your loved ones.
This might be counter-intuitive. The reason is that retirement is long – say 30 years. Stocks usually recover from declines. Holding cash for 30 years means you lose out on a lot of income, plus the loss of purchasing power due to inflation.
Ed’s advice: Forget the cash buffer. Just choose your investment allocation and stick with it.
New Retirement Rules of Thumb (supported by history)
The 5 rules of thumb about retirement income on which the advice for most seniors is based are not supported by history. What new rules of thumb give you the maximum reliable retirement income?
- Equities are safer. Don’t feel that you should invest more conservatively just because you are retired. Retirement is for 30+ years. Consider keeping the same allocation you had before retiring. Equities are taxed at much lower rates than bonds & GICs, as well.
- Replace the “4% Rule” with “2.5% +.2% for every 10% in stocks Rule”. Instead of withdrawing 4% of your retirement assets, use the formula if you have less than 70% in stocks. Fixed income is lower income. The more conservatively you invest, the lower your retirement income should be.
- The “Age Rule” works with a “3% Rule”. There is nothing wrong with investing conservatively with the Age Rule, but then reduce your retirement income to withdraw only 3% of your investments each year.
- Be smart about your risk tolerance. Invest with the highest amount in stocks that is within your risk tolerance. The more conservatively you invest, the more likely you will run out of money (at any withdrawal amount). Get educated on stock and bond market history, so you have an accurate picture of risks and returns.
- Inflation is huge. Inflation typically makes the cost of living triple during your retirement. You need a rising income, not a fixed income. Inflation kills bonds, but not stocks.
- It is safer NOT to hold cash. Holding cash does not protect you and may increase your risk of running out of money. It almost definitely means you die with a smaller estate.
- You can have all 3 – Higher income, lower risk of running out of money, and less tax. You can have a higher reliable income, if you can manage your income effectively or are working with a financial planner who knows how to manage it effectively.
For an example of these principles, read the story of John & Jennifer:
Can we retire now? Ed Rempel tests retirement income rules of thumb
Ed
1 cFIREsim – Data from S&P500, Barclay’s US bond index & US inflation.
2 Prof. Jeremy Siegel in his classic “Stocks for the Long Run”.
Planning With Ed
Ed Rempel has helped thousands of Canadians become financially secure. He is a fee-for-service financial planner, tax accountant, expert in many tax & investment strategies, and a popular and passionate blogger.
Ed has a unique understanding of how to be successful financially based on extensive real-life experience, having written nearly 1,000 comprehensive personal financial plans.
The “Planning with Ed” experience is about your life, not just money. Your Financial Plan is the GPS for your life.
Get your plan! Become financially secure and free to live the life you want.
Hey. Very intelligent write up!
I am confused about one thing. When you have 100% equities and no bonds/cash, how do you survive a downturn/crash/recession? Where do you get the money to fund expenses without ruining your retirement portfolio? Do you still pull out 4%? Or do you pull out 2% (Still that is 20k reduced from 40k, how will this cover expenses?)
Thanks
Hi Roy,
Good question. We are seeing a lot of people in the FIRE movement able to retire much earlier, even in their 40s or occassionally 30s.
The longer the time period, the more predictable returns have been – especially for stocks. In the chart on standard deviation (risk) by time period, you can see that the risk of stocks drops quite a bit from 20-year periods to 30-year periods, but bonds and cash only drop a bit. Longer time periods should show even lower risk for stocks and a bit lower for bonds & cash.
In the vast majority of 30-year retirements with 100% equity portfolios, you are way up at the end and not in any risk of running out of money. Extending it another couple decades would put you even farterh ahead.
I have not run tests for periods longer than 30 years. There may be quirks in such studies related to our limited data. We have about 150 years of detailed reliable data. Any 50-60 year period would include more than 1/3 of all the data. There were really 2 difficult times in history (1970s and 1930s), so studies of longer time periods would mean nearly all periods would include one of the difficult times. Studies of longer term periods may give results more related to data quirks than real insight.
Ed
Hi,
What if someone chooses to retire early. How does the success rate react to portfolio allocation (ie the graph included) over a 50-60 year period ?
Hi Noel,
It is usually the low-return investments that are the most highly taxed, such as bonds and dividend stocks.
Ed
Enjoyed reading your advice on the different issues arising at retirement. I believe that risk taking depends on a large extent to the portfolio value .i.e if I have $1 million my risk taking can be higher but why would I take a risk if I have to pay back a lot in tax?. I will soon be celebrating my 80th b’day.
[…] and tax accountant Ed Rempel challenges all these conventional retirement strategies in this article and offers alternative ones. […]
[…] fiscaliste Ed Rempel remet tous ces concepts conventionnels de la retraite en question dans cet article et offre des stratégies alternatives. […]
[…] investments in a non-registered account, and systematically sell a bit each month (following the “4% rule”) the cash flow is either a return of capital (your original principal) and therefore not taxable, […]
Hi,
Accumulation-Dynamic decumulation is based off percentage withdrawal of the current fund, versus initial fund. Thus the method can be used in combination with a constant percentage or using the variable percentage method depending on the bequest motive. Using a percentage of current fund means that there’s no risk of ruin of the fund or sequence of return risk, but it does bring risk of adverse withdrawal in the future.
The reserve fund is based of 25% of the total fund invested in the same way as the regular fund. The remaining is left in the regular fund in which you withdraw each year. Ex: Withdrawal rate =5%; Smoothed fund value= 1million; Reserve fund= 250k; initial withdrawal = 50k = 5%/75%*750k. If the fund does good, the reserve is left to grow as no withdrawal is done on it. If the regular withdrawal goes below 90% of previous withdrawal level, the reserve covers the missing portion. Ex: previous max withdrawal = 50k, current withdrawal would be 40k, we would withdraw 10k from the reserve fund. Often the reserve mechanism lead to following the Market increase until market crash and then followed by a long period of level withdrawal. The reserve is protected against depletion by putting a maximum withdrawal rate on it of 20%. In the case of very low economic, this helps ease the pain to adjust slowly from 90% of previous level to available fund withdrawal rate, instead of the crash and burn technique of Safe Withdrawal.
At last, as the name suggest, accumulation-dynamic decumulation also look at the accumulation phase to determine the initial withdrawal. The smoothing technique and a cap is apply on the “projected withdrawal” so that a huge market gains prior to retirement will be met with a more conservative. This way, the method mitigate the retirement date risk.
Hi Patrice,
Tell me more about your accumulation-dynamic decumulation method. How does it work, Patrice? I’m always trying to study it more.
I tested all kinds of methods. I found a variable withdrawal, cap, and smoothing can all work, but I did not find success with reserves.
The main focus of my study was on choosing a withdrawal amount and increasing it every year by inflation. What if we managed the withdrawal, instead of increasing it every year?
I looked at a wide variety of ways to manage your retirement income, including only increasing by inflation in good years, having some type of cap and floor on the withdrawals, reducing withdrawals by some percent if they get too high, etc. I compared my best strategies with various studies, including several actuarial studies and some advisor formulas, such as the Guyton-Klinger strategy and the Hebeler Autopilot.
I tried to find the best formula to manage retirement income to allow a higher and safer income.
The options are complex, but I found there are effective methods that had 100% success in history with withdrawal rates of 5% and even 6% of your investments. You should be careful with these higher methods, since you will have to manage your income effectively.
Ed
Hi Ed,
Like you I figure that you are better off with a higher equity.
I’ve created a new decumulation method that might interest you. I go basically down the path variable withdrawal, but including an increase cap, some smoothing and more importantly a reserve. My conclusion was that you could increase substantially the withdrawal this way while keeping much of the downward risk of volatile withdrawal at bay.
Please let me know what you think of the accumulation-dynamic decumulation method.
[…] Is Typical Retirement Advice Good Advice? – Testing Retirement Rules of Thumb […]
Hi Jonathan,
Good idea. Perhaps that can be a future article.
One big myth about gold is that it is an inflation hedge. The last time I looked at the correlation, gold and inflation have a correlation of -.3 over the last 3,5,10 and 20-year periods.
This means that gold usually goes DOWN with inflation.
For example, this is part of why gold went to $2,000/oz. Inflation at the time was very low and there were expectations of deflation. Gold has since dropped a lot as inflation expectations have risen.
I believe this is because higher interest rates make gold less attractive. Interest rates are usually higher during times of higher inflation.
Gold has gone up during some stock market declines, not nearly all. It seems to rise in periods when there is more speculation of a complete collapse of our entire system.
I would expect the results to show that gold has probably never, or almost never, been a protection over a 30-year retirement.
Ed
HI Dale,
I agree. Having 70% in equities has historically been as reliable as 100% equities with a 4% withdrawal.
It has not done as well with higher withdrawal rates. If your withdrawal rate creaps up to 5% or 6%, then 100% equities has been more reliable in history.
Ed
Hi Alain,
I agree with you, in general. How much you invest in stocks vs. bonds is partly a risk tolerance issue and partly based on the rate of return you need to achieve your life goals.
In practice, when I have clients that are on track to have more than they need for their financial independence (retirement) goal, we usually keep a safety buffer and then discuss the specific purpose for the excess.
For example, if your financial independence number is $1 million and you are on track for more, it is good to have a safety margin of 10-20% or more in case something goes wrong in the future. You are more likely to have an unexpected large expense than receive an unexpected large amounnt of money.
For this reason, it’s probably wise to build up to about $1.2 million in your long-term portfolio, which could be 100% equities.
If you are far ahead, it is good to make a conscious decision with what to do with the excess. If you are on track to have $2 million, and you only need $1 million to provide the lifestyle you want for life, what do you want to do with the other million? You can afford to enjoy your life more and spend more. You could use it to make a difference in the world by giving it away or donating it.
“Not to decide is to decide.” If you don’t make a conscious decision for the exta million, it will likely grow and you will leave a larger estate. Is this what you want? If you die with a multi-million estate, does that mean you missed out on enjoying things during your life?
What is the most fun or meaningful use for your extra money?
Ed
Hi Rob,
Holding some GICs or bonds is about risk tolerance. You lose less in a down year. If you need it to be within your risk tolerance, then it makes sense. Having only 10% in fixed income should not require a lower withdrawal rate either.
Over a 30-year retirement, though, holding any amount in cash, GICs or bonds essentially never been a benefit. 88% of down markets have recovered fully in 1 or 2 years in history.
You should assume a lower long-term average rate of return. I would suggest about .6%/year lower with 10% in fixed income. For example, with 10% equities, it can be reasonable to expect 8%/year return over time. With 10% fixed income, 7.4% would be more reasonable.
Ed
Very intersting. Another interesting mythbusting endeavour would be to investigate varying percentage in gold to confirm or dispel its “safe haven” halo.
The Balanced Growth model appears to work quite well – 70% or so in well diversified lower volatility stocks with some international diversification. I have no problem with the 4% rule. It worked easily through the last two recessions and major market corrections, Two very trying times for investors. The US Vanguard managed portfolios made it through with ease on the 4% plus rule. I had put that in a Seeking Alpha article.
It does work, or did work, if the portfolio is set up correctly, but in simple fashion.
Dale @cutthecrapinvesting.com
Great job. It’s nice to see that the standard deviation for stocks is the same as for bonds after 20 years. I knew this intuitively but is nice to see the research by someone else.
My take is to be 100% in stocks until you hit your Financial Independence number with the 4% rule. After that, if a person continues contributing, then they can put the extra money in bonds, and play it safe or go the other extreme and put it in more speculative investments.
For example, if your FI number is one million, then, any money in excess of one million can be placed in bonds.
Also, it’s important to consider the amount of money. Someone with 10 million can afford to put 100% into bonds and have a very nice life and someone with $100,000 cannot afford to have any of his money in bonds.
Hi – I liked just about everything you’ve said, but am stuck on the cash reserves or at least doing a 3-5 yr GIC ladder. Your table shows that cash does not leave you richer but it seems that you are studying over a long period of time – which is the mindset for investing but remember this is to protect oneself for a very short period of time; just until the market recovers. It would seem that these aren’t apples to apples comparisons then as the drop in market you need to ride out is for a fairly short period of time, not 30 years. Particularly if that drop comes very soon after retirement. Also my GIC ladder wouldn’t represent too much of the overall portfolio (less than 10%) so I don;t feel I’d be loosing a tonne of the gains. What say you?