New Thoughts on the 4% Retirement Withdrawal "Rule"

A standard rule of thumb has been that retirees can safely withdraw 4% or 5% a year from their investments and in almost all cases have enough savings to last the rest of their lives. So if you had $1 million saved for retirement, you could take out $40,000 to $50,000 per year for the rest of your life. Supplement this with Social Security and other earnings, and you'd probably be doing quite well.

But perhaps the 4% or 5% rule isn't really that accurate and can't be counted upon. Here's what the NY Times says about a new analysis on this subject:

The inexact science of withdrawal rates was featured in an analysis published by Vanguard Group in November. While a withdrawal strategy "is important, the key ingredient in a long-term spending plan is flexibility," the report said.

If investors are relying on either gains in the stock market or bond-market yields to make their money last, "then investors must either accept continuous, relatively smaller changes in spending or else run the risk of having to make abrupt and significantly larger adjustments later," the report said.

Rather than take out a steady 4% or 5%, the Vanguard report suggests many investors would generally stand a better chance of not running out of money were they to adopt a strategy where the percentage of withdrawals was designed to rise and fall between 2.5% and 5% of the prior year-end portfolio, depending on the market's ups and downs. In short: After a good year, take out more, and following a bad year, less.

However, one challenge this approach presents is that a 2.5% to 5% band represents a huge amount of variability in income for a retiree relying on savings to help pay the bills.

Uh, yeah. There's a HUGE difference between taking out $25,000 a year and taking out $50,000 -- especially when you don't have many other income options and fixed expenses.

The piece goes on to tell how the 4%/5% "rule" won't work in many situations these days:

Investors had an overall success rate of 95% at a 4% withdrawal rate adjusted over time for inflation. But that success rate fell to 76% for those who retired when stocks were in the top 20% of the historical valuation range.

The numbers were starker for a 5% withdrawal rate. For retirees starting out when stocks are in the highest 20% of valuations, the probability of success plunged to 41%. Only when stocks were in the bottom 40% of their valuations did retirees stand a better-than-90% chance of not running out of money.

The bad news for today's retirees: on this metric, stocks today are in the top 20% of their historical valuations.

So what's the solution? A few options:

• Save more. Don't retire until you have enough saved so you only need to withdraw 2.5% per year to meet your needs.
• Get additional sources of income while in retirement. The most common here is working longer, but you can't always count on doing this.
• Be able to cut expenses. If some of your retirement costs are variable, you can cut them when income goes down so you'll be fine.

So it looks like the 4%/5% "rule" isn't the standard any longer. For me, it never really was. I have always been working on saving much more than what the "rules" say (I'm just conservative with my finances like that, as you know.)

How about you? What are your retirement withdrawal plans?

Comments

[DIY Investor]

It isn't just about the percentage. Retirees need a plan for withdrawals. One that has never failed in actual cases is to have 1 years worth of payments and to structure the portfolio to throw off 60% in income needs (2.4% yield) to replenish funds. This enable the retiree to ride out down turns.
Do the analysis and see what happens to your 76% probability of success ratio. Drawing down the portfolio is a lot different from building it up. You need to understand the potential impact of reverse dollar cost averaging and have a plan to attack it.

[Mike Hunt]

If you do a reverse annuity then you know the payments are fixed and you only need count on inflation... but with rates being held as low as they are compared to historical levels this is a bad time to get value for your annuity in my opinion.

I'm planning on a 1-2% level of earning of my retirement portfolio just to be safe.

One thing is for certain- nothing is guaranteed for the future. Have a well laid plan with lots of contingency options built in!

-Mike

[Venkat]

I think kids taking care of their parents will greatly improve one's retirement(I took care of my parents until they passed away). I understand it is a foreign notion to most of the Americans(it always me me here in America--Just an observation).

After all, Parents work hard for us when we are kids, I think it should only be natural for kids to care of their parents(I also realize it is not possible for everybody to care of their Parents based on their Individual Circumstances).

In order for the above to happen, our social thinking fabric has to change.

[ryan]

Can't you earn more than 2.5% a year on the money you still have invested?

So say you plan to withdraw 2.5% a year in retirement. The rest of your money is still invested and should be earning something. So really if you're able to take out just 2.5% or maybe even 5%, aren't you replenishing each year on the interest gained from what's left in your portfolio?

[Everyday Tips]

Ideally, I won't count on any withdrawal. I have been putting more thought into fixed annuities that I would buy right before retirement so I have a guaranteed income flow, in addition to having money set aside.

I think 4-5 percent would be way too high.

[tom]

That is about 40 years away, but I plan to take out what I need. Some years it will be 2.5%, some will be 7% (maybe we want to go on a big trip).

The average will hover around 4%.

The other side to this is that your money is not just sitting in an interest free savings account. It's probably sitting in a portfolio comprised of mostly bond funds. It's also probably earning a 5% annual return on average. Basic math says that you'll never run out of money.

Frankly we should all be spending more in retirement, or retiring earlier. I definitely want to leave something to my kids, but, hopefully, they will be old, well off and possibly retired by the time I die.

[jp]

This rule of thumb is only useful in determining a good starting point for retirement. It's not realistic to use for much else. If you continually withdraw 5% of the prior year-end value, you will obviously never run out of money. You are always saving 95% of your assets regardless of age. On the other hand, continually withdrawing 5% of your initial retirement portfolio without any regard for your current assets is just silly. For example, a person who retires at 65 with $1 million ($50,000 withdraw rate) might be able to grow that portfolio to $3 million by age 80. An 80 year-old can obviously take more than $50,000 an year from a $3 million portfolio without a great risk of ruin.

[Marcin]

I'm always amazed how people don't realize that stocks market can significantly go down in very short time (like 2008 crash). You can not expect to be able to live out of 4% of your investments when all of the sudden $40k/year can be 30% of your portfolio. Current market is up only because of trillions from government. How many times will we be able to afford trillions during upcoming crashes. We got the market up, be the same money will have to be repaid in taxes soon (of course not by those who got the money from bailouts ;) )

[Apex]

Once you reach required minimum distributions (age 70) you will be taking out 4% a year and more whether you want to or not. At age 70 the required distribution is 3.65%. At age 78 its 5%. And it keeps going up from there.

In order to not do reverse dollar cost averaging as someone above mentioned, it is important that a retirement portfolio that is being drawn upon have a considerable amount of funds in very stable fixed income investments. I have not thought about how much considerable is, but its definitely more than 10%. Old Limey might be an exception due to the mountain of gold he is sitting on :) . The purpose of this is if the market is off drastically you can take your distributions mostly out of the fixed income portions giving your investment portions time to recover so you don't sell low. When the market is roaring, take your distributions out of the investment portion and let your fixed income assets sit. This is also a poor man's way of doing asset reallocation but it will serve its purpose rather well.

[JimL]

One option is to go for dividend paying stocks and try to seek a 5% payout. The key is having a good mix of companies with a strong cash flow. If the market drops, you still have the dividend flow and you don't need to liquidate shares at a low price to live off of.

[Apex]

@JimL,

That seems like it works except that when the market has a steep correction, most high dividend paying stocks slash their dividend. If this strategy was followed this last decade it would have had you invested in many financial and bank stocks. They paid some of the higher dividends. Many of those stocks lost 80-90% (some lost 100%) and their dividends were often discontinued or cut to a penny. If the stock of the company that pays a dividend gets hit hard, the dividend will be slashed.

In the early 1990s IBM was having trouble as a blue chip bell whether. The stock lost a large portion of its value but IBM tried not to touch the dividend to make investors feel the company was still safe. But eventually they slashed the dividend too. You can't justify paying out a high dividend when you can't bring in the profits. Where does the dividend come from if the profits have gone away?

[FMF]

JimL --

Good thoughts. I have been considering that line of thinking too...

[jp]

Apex,

Withdrawing 4% from a 401K is not the same as spending 4% of your assets. Required minimum distributions do not have to be spent.

[Apex]

@jp,

That is true. However this post was about withdrawals. Also if you withdraw the money in a down market and don't spend it you have still taken the investment loss hit. I suppose you could put it back in to the same investment in a taxable account. I doubt very many people do that.

[ryan]

Apex...there are companies that have increased their dividend every year no matter what. I think it's Dividend Aristocrat or something to that effect. If they don't increase each year (or at least maintain), they are removed from the list.

[Apex]

@ryan,

That's a good list and a good track record. There are probably better ways to do the dividend plan than others. A good diversification across industries is of course helpful. I am just trying to point out that this plan can have cracks. Would IBM and AT&T have been on that list in years past. I suspect they probably were. But both ended up slashing their dividends badly when they struggled and both were at one point in time thought to be nearly untouchable.

Those who survived the past recession unscathed might be the ones hit hard in the next. No one knows. The future is uncertain and dividends are not sacred.

[FMF]

Apex (and others) --

You could also consider a "Dogs of the Dow" or modified DOTD strategy -- dividends plus some (good?) potential for upside...

[Rick Francis]

I've still got a lot of time to go before retirement, but it seems that it would be a good idea to have multiple accounts for different purposes:
To cover minimal living expenses (housing, food, insurance, etc. but no vacations, eating out etc.) use very safe investments (i.e. Social security, TIPS, CD ladder or an annuity). Since I expect to not have a mortgage when I retire SS may be enough.

For discretionary spending an investment portfolio that has some risk. I’m not sure the exact split but possibly 50% stock /50% bond index funds. Take money from the portfolio based on its real annual returns. That insures you will never run out of money- you just have to be disciplined enough not to spend all of the discretionary spending in the good years so that you have some discretionary spending in the bad years.

For emergencies a savings account with a few months worth of living expenses. This shouldn’t need to be outrageously large as long as you have reasonable insurance coverage to hedge against catastrophic costs.

-Rick Francis

[Apex]

@ryan,

As a follow up I found the dividend aristocrats list from 2007 and compared it to 2011.

2007 List - http://seekingalpha.com/article/43046-s-p-s-dividend-aristocrats-ranked-by-yield
2011 List - http://seekingalpha.com/article/243143-dividend-aristocrat-list-for-2011

In 2007 there were 59 companies on the list. In 2011 there are only 42 and some of those were not on the 2007 list. This means that more than 1/3 of them fell out of the list from 4 years ago. This is what I mean by recessions and stock collapses slashing dividends regardless of their history.

Secondly these are not necessarily high paying dividend stocks. in 2007 only 4 of the 59 paid 5%, another 7 paid in the 4% range, another 5 paid in the 3% range. Most paid around the 2% area and 21 of them paid less than 2% with 7 of them paying less than 1%. The average of the bunch would have been less than 2.5%. If you went for the high dividend paying stocks of this list guess where you were? Banks. They mostly went to zero. If you left out the high dividend stocks and banks thinking they might be risky you would get a low dividend payout.

The other thing about a list like this is its a backwards looking list. It's like buying mutual funds because their 10 year track records has been good. Banks were good for decades, until they weren't. Knowing that they had been good in the past doesn't tell you how well prepared they are for the future. It might just tell you they have been lucky. These lists promote that they have outperformed other indexes over the past X years. But that is after you know they are on the list. That's cheating. The 2011 list I posted above brags that it has outperformed the s&p500 5% to 1% over the last 5 years. But 5 years ago you didn't have the 2011 list. You had the 2006 list. How did the 2006 list perform over the last 5 years. I can tell you matter of factly by looking at the 2007 list that it performed horribly, both in terms of dividend and stock performance.

As I said, there may be better ways to do the dividend thing than others but its tricky.

[Deserat]

From a blog by Doug Nordman (http://the-military-guide.com/):

"One example of that vigilance and flexibility is Bob Clyatt’s “4%/95%” approach from the book “Work Less, Live More”. Bob’s system is one of the very few “variable withdrawal rate” approaches to retirement spending. It looks at the portfolio every year and takes a straight 4%– no adjustment for inflation because the 4% is calculated at the beginning of every year. As you can imagine, this method is wildly popular if the stock market’s rise exceeds inflation. It’s not so cheery if the market is flat and inflation is rising. Even worse, spending has to be cut when a bear market inevitably rears its fanged head– which is where the “95%” kicks in. If next year’s 4% calculation is less than 95% of this year’s 4% calculation, then next year the ER has the option to take 95% of this year’s 4%. This is a temporarily higher withdrawal rate, but it still temporarily reduces spending to minimize damage to the portfolio. As the bear market ends and the portfolio recovers, ER spending can rise right along with it at the next 4% calculation."

Clyatt's book is quite a good one - Doug Nordman's book is good, too (disclaimer, I contributed to Doug's book on military early retirement)

[Derrik Hubbard, CFP]

I would suggest that those nearing or in retirement consider an "absolute return" strategy for their assets, one seeking positive returns in all market environments.

Too much of an allocation into stocks is too risky for the retiree, and too much of an allocation to bonds in this interest rate environment could also be disastrous.

Derrik Hubbard, CFP

[Old Limey]

I am now 76 and for 2010 my minimum required distribution was 4.6%. I have year-end witholdings for State and Federal taxes and the remainder is transferred into our Trust account which is of course taxable.

I am invested conservatively and have CDs that average out at 4.93% but they were purchased in October 2008 when I was able to buy CDs yielding between 4.8% and 5.15%. It's an altogether different story these days. With some of my CDs that were redeemed by the FDIC because of bank failures I have that money invested between 3 income funds that are currently yielding 4.7%, 7.8%, and 8.2% but their value is of course subject to market fluctuations. So far, so good, they have capital gains, but I watch them closely and would be out of them quickly if they started slipping much in value.

Fortunately we have no debts and can live well on our pensions and SS checks so it's not a problem for us. Most of the next generation however don't have pensions to look forward to. Let's also hope that Medicare continues as is for those that are now under 55, it would be a real hardship for them if it ends up as a voucher program.

[JimL]

Apex,

I was going to suggest what FMF said. The key is to have a good mix (consumer, utility, healthcare, etc) and as said before, watch for those with good cash flow.

[JimL]

Old Limey,

It would be great if we could continue with Medicare as is, but the math just doesn't work. First, we have demographics working against us as we have an aging population that is also living longer. Secondly, the American lifestyle is out of control as we have a severe diabetes issue as well as increasing cardiovascular disease that are both being driven by rising obesity rates. Did you know that the average person over the age 60 that is seeking medical care is current on 6 different medications? Based on current trends, a person born today would have to cover a 122% increase in the Medicare payroll tax and accept a 51% cut in services or respective payments.

We often put the blame on doctors fees, but they are usually losing money on government reimbursement, which is offset by private pay/private insurance. The other scapegoat is pharmaceutical companies. However, pharmaceuticals is a piece of the pie compared to other categories and the use of generics is now almost 70% of the current unit volume.

The potential fix is for people to eventually take on more burden to incentivize them to live a healthier lifestyle.

[Ginger]

When I plan, I assume a 3% withdrawal rate. People say, if you want the money in the next five years don't put it in stocks, yet retirees do. I plan to have my yearly expenses in cash and then for year 2-5 have the money in a bond fund. That way I can not worry about money when the stock market crashes. Also, I plan to have real estate investments which should also help.