Retirement Withdrawal Strategies

New Reader? Get free regular updates from Can I Retire Yet? on saving, investing, retiring, and retirement income. New articles about twice/month. Join more than 11,000 subscribers. Unsubscribe at any time:

You’ve worked and saved for years, most of a career. Maybe you’re looking at a traditional retirement in your 60’s, or you’re one of the fortunate few who are able to consider earlier retirement. Either way, you’ve amassed a sizable nest egg. Your assets are likely well into seven digits, unless you live very frugally, or can rely on a guaranteed pension. That sounds like a formidable sum. You know it will last for years. But how long, exactly? And how should you start living off of it, making withdrawals, to guarantee it won’t run out?

Most of us without pensions will face these questions in our retirement. How do we safely withdraw from a lump sum over time so that it lasts as long as we do? We know there is some baseline of living expenses that we must support. But then there are a host of discretionary and one-time expenses on our plate — travel, maintenance, new vehicles, emergencies. When is it OK to spend this money, and when might it be damaging to the long-term survival of our portfolio? Is there a system, a technique, an investment, a product, that will keep us safe?

Below I’ll do a high-level survey of all the many ways to withdraw from your nest egg in retirement. I’ll cover every method or strategy that I’m aware of, from a general perspective. I’ll also discuss what I’m doing, at this stage of my retirement. I won’t get into hard numbers or simulations just yet. We’ll leave that kind of analysis to future posts, where I’ll explore new research and what it means for you, and me, in more detail.

By the end of this post you’ll have a good, general sense of your retirement withdrawal options — both their positives and negatives. And you’ll understand why there is no perfect answer to this question. Every solution has its strengths, and its potential weaknesses. Only by understanding the possible approaches, then mixing them together into a personal solution, while remaining flexible, and still accepting some risk, will you be able to move forward and enjoy your retirement.

Fixed Withdrawals

Let’s start with one of the simplest and therefore most popular withdrawal methods. That is withdrawing a fixed amount from your portfolio on a periodic basis. Typically this is implemented by also adjusting for inflation annually, so the nominal amount actually grows over time, but remains constant in real terms. In other words, you are maintaining the same lifestyle from year to year.

If the amount you start with, in year one of your retirement, is 4% of your portfolio, then this is the classic, but possibly obsolete, 4% Rule.

The advantages of this withdrawal method are that it is relatively simple to implement. And it has been studied exhaustively so you can easily find statistics on the survival probabilities for your portfolio, given a time span and asset allocation.

And this strategy is perfectly predictable — to a point. Your lifestyle is “locked in.” You know much you have to spend each year. Until your money runs out.

That is the chief drawback of the fixed withdrawal amount. Nobody can say for sure how long your money will last. Yes, studies based on historical data show it might last for 30 years. But history may not repeat. And, unless you implement some kind of annual review, you will have no flexibility to make your money last longer, if needed. Nor will you be able to deal with emergency expenses, or enjoy splurges, if your portfolio is doing well.

Variable Withdrawals

If you don’t withdraw a fixed amount from your portfolio every year, then it stands to reason you’ll be withdrawing a variable amount. But there are many formulas for computing that fluctuating withdrawal. Let’s review them.

You could withdraw a fixed percentage of your portfolio every year. Say you chose to withdraw exactly 5% of your portfolio every January, regardless of its current value or how the market had performed. This is often called an endowment withdrawal approach. The great advantage of this is that it automatically builds some flexibility into your withdrawals as a function of market performance. If the market is up, your fixed percentage will be a larger sum. If the market is down, it will be smaller. Even better, you will never run out of money! Because you are always withdrawing some percent of your portfolio, it will never be wiped out, mathematically speaking. That sounds great until you look at the downside: your portfolio could get very small! Your lifestyle will fluctuate, perhaps dramatically, over the long term. Lastly, although this withdrawal strategy is variable, it isn’t truly flexible, because the market, not you, controls the size of annual withdrawals.

(For a different approach to a market-based withdrawal percentage, there are market valuation-based strategies. These advocate that you actually withdraw less when markets are high, because the projected returns going forward will be lower. There are pages of debate on this topic. For one view among many, see leading researcher Wade Pfau who concludes in one article that valuations “may not really be of central importance when building an overall retirement income strategy.”)

Another approach to computing a variable withdrawal sum every year has been gaining ground recently, and that is to base the withdrawal on your life expectancy. In its simplest form, each year you would withdraw one divided by your remaining life-expectancy in years. (For a simple source of life expectancy data, see the IRS RMD tables). So if your age was such that your life expectancy was 30 years, you’d withdraw 1/30 or about 3.3% in the current year. This approach has some intuitive appeal. You are dividing your money up evenly according to how many years you expect to remain living. And that expectation is updated annually based on statistics. Again, you will never run out of money! But again, there is no guarantee exactly how much money you’ll have in your last year. However, if you consider how the denominator in that formula works, along with the potential for asset growth, you’ll likely wind up with smaller withdrawals in early retirement, and possibly large withdrawals later. That would be a potential criticism of this withdrawal method: Instead of “front loading” withdrawals so that you spend more in your early years when you have a greater chance of being alive, and are healthy to enjoy it, you are likely to “back load” the withdrawal process, saving too much until the end.

If you want to allow for more generous withdrawals up front, plus potential flexibility, yet institute some safety measures, you can go with a variable withdrawal strategy consisting of a fixed percentage with some bounds or guardrails. There are some more modern versions of this strategy, but one of the first and simplest approaches came from Bob Clyatt in his groundbreaking book on early retirement Work Less, Live More. He calls his approach a better Safe Withdrawal Method with “The 95% Rule.” You start by calculating 4% of your portfolio value every year. Then, to accommodate bad market years, you can withdraw either that 4% amount, or 95% of the amount you withdrew the previous year, whichever is larger. That means you never have to cut your lifestyle more than 5% in any given year. Clyatt reports that there is minimal impact on portfolio success rates (generally just a few percentage points) from using his 95% rule. So you are able to “smooth” changes in your lifestyle, without taking on much additional risk of running out of money.

Annuity Payments

What if you want even more retirement guarantees? On the surface, annuities appear to solve almost all the problems of fixed and variable withdrawals. Almost.

With an annuity, you hand an insurance company some or all of your retirement portfolio. In exchange they give you income payments for life. That eliminates the possibility of outliving your assets. That’s a very good thing. Of course that guarantee depends on the solvency of the insurance company and its backing state guaranty association, both of which ultimately depend on the markets and economy. That’s the first flaw in the annuity’s glossy exterior.

Annuities get high points for consistency. You are party to a contractual obligation that you will receive the same amount, year after year. But that’s also the chief problem with an annuity. It’s inflexible. If you die early, you will likely have left quite a bit of money on the table. If you have an emergency and need a lump sum of money, you probably can’t get it. Only if you live a long time, and your expenses stay relatively constant, does the annuity solution come into its own.

But let’s take a closer look at those “lifetime” payments. We’re talking two, three, four or more decades of cash flow. Is it inflation adjusted? Inflation can be a big deal over such time spans. And, today at least, true inflation-adjusted annuities are hard to find and feature minuscule payouts. (Looking at my latest inflation-adjusted annuity quote, I’d rather take my chances in the markets.)

The final drawback to an annuity is that you don’t get to keep your principal. So there is no option for emergency cash or leaving an inheritance with that particular portion of your assets.

Flexible Capital Preservation

We’ve discussed fixed and variable withdrawals, and annuity payments, including the positives and significant negatives of each. These are all relatively static, idealized, or inflexible systems for withdrawing from assets in retirement. They make great case studies for performing research, running simulations, and publishing scholarly articles. The problem is that to be simulated they must all make assumptions about the future of the economy and your personal financial situation. Assumptions that may not apply, or may not come to pass.

Now let me discuss what I think most people actually tend to do in retirement, at least early retirement. And, this is also what I’m doing, until I see better ideas and a more complete picture.

I will call this “Flexible Capital Preservation” — living off interest, dividends, growth, and some part-time work income, all with an eye to preserving assets. This is, in my view, the only sensible strategy for early retirement, where there are still decades of uncertainty yawning in front of you. If you retire in your 50’s or earlier, with no inflation-adjusted pension in sight, it is simply unwise to draw down your retirement assets in any significant way. The goal should be to preserve (though not necessarily grow) your net worth, keep your powder dry, and not burn any bridges, until you are much further down the road.

How does this work in practice? Well, if your assets are large enough, or the markets are strong enough, you can afford to spend only your annual dividends and growth each year. If that isn’t the case, then you will need to work part-time, possibly leveraging your relative financial independence to create a lifestyle business that supplements your investment income while in no way detracting from your quality of life.

Flexible Withdrawals

So how exactly do you go about withdrawing from investments when taking this flexible approach? I’ve been doing this now for nearly three years, in a generally rising market. That tailwind has made my financial life much easier. But, regardless of the state of the economy, I would take a total return approach that might be characterized as “just in time,” or “active safe withdrawals” (as opposed to active investing). In other words, once or twice a year, when I need money, I sell some of my most appreciated assets.

Isn’t that a bit like market timing, in reverse? Maybe, maybe not. But a critical distinction is this: I’m selling in response to my own routine income needs, and not in response to market events.

But isn’t that still risky? What if the market tanks all of sudden? How do you avoid being forced to sell your assets at a loss?

This is where asset allocation comes into play. Some have formalized this as a buckets system — keeping less volatile assets like cash and bonds on hand for near-term living expenses. One prominent buckets strategy has been discredited, but the concept is intuitive to many and still useful for discussion purposes. In truth “buckets” are nothing more than another way to manage your asset allocation.

For me, it makes no sense to live off your conservative cash and bond buckets when stock markets are up. That’s like dipping into the storehouse when there is fresh, healthy grain available in the fields. Much better to be selling volatile equity assets when they are in favor, and to preserve your safe “buckets” for the bad times.

When, those bad times come, as they inevitably will, then the flexible withdrawal approach begins dipping into the safe cash and bond buckets that have been set aside. Ours are large enough to live off for a decade or more, before we might even have to consider spending more volatile equity assets at a loss.

There is even new research suggesting that it may actually be better to spend bonds first, producing a rising equity allocation for the long term. So, if required by economic events, it is probably OK to spend down your conservative buckets to very low levels.

For optimal retirement income, you must take into account your economic environment. A flexible approach is best, validated by both intuition and the laboratory.

Hybrid Strategies

Every retirement withdrawal technique described above has drawbacks. Some can’t guarantee that you won’t run out of money: you could die broke. Some can’t guarantee your lifestyle: you could be living out of a trailer and eating cereal in your later years. Some don’t guarantee principal: you might not have any money available for emergencies, and you might not have any left to pass on to your heirs. Some require active involvement and financial management: you or a spouse must be able and interested.

Despite all those drawbacks, the biggest problem with most of these withdrawal systems is precisely that they are an artificial system. Ask yourself if it is really possible to maintain any “system” consistently for the 20-40 year time span of a modern retirement? Has there been any significant process that you undertake today precisely as you did decades ago? Whether it’s fixing your meals, maintaining your vehicles, planning your vacations, or managing your investments, there are ceaseless changes — both subtle and vast — over time.

The way I invest now is quite different from how I invested just 10-15 years ago. There is more and better information available. There are new, better, lower-cost investments available. My understanding and needs are different. Try as we might to stick with any financial system, things change.

Even annuities, which are arguably the most “reliable” retirement withdrawal system, suffer from change. Their payouts may seem predictable, but the world changes around them. Inflation could take off, seriously eroding their buying power. You could die early, or have serious emergency expenses — scenarios that are negative for annuities.

That’s why, when all is said and done, I believe most of us are going to construct a flexible, “hybrid” system for living off our assets in retirement. We’ll pick and choose from the options, combining them in an attempt to harvest most of the benefits, while minimizing the liabilities and preserving our flexibility. In many cases we’ll try to create a guaranteed, inflation-adjusted income floor plus an upside of investment growth for recreation, emergencies, and legacy.

Though research is underway to compute the “optimal” formula for this hybrid retirement income stream, I remain somewhat pessimistic that there will ever be a perfect, turnkey, “just tell me what to do,” solution. All the same, I know that many of us will find solutions that are “good enough” to enjoy our retirement safely and comfortably.

My own evolving strategy for retirement income will be the topic of future posts. If you have ideas or lessons learned for your own ideal retirement income solution, I’d love to hear about them below….

* * *

So, that’s a wrap for 2013. It’s been an eventful year here at Can I Retire Yet? — publishing my first book, selling our house, downsizing, and moving west. I’ll be taking the rest of the month off to enjoy the holiday season. And I hope you are able to do some of the same. Thanks for all the support this year, and be sure to check back here in the New Year for more posts on planning and living comfortably in retirement!


  1. Excellent post, Darrow! This is a topic that really interests me so I look forward to your future articles.

    My plan is quite similar to your Flexible Capital Preservation approach but is a bit more risky in that I plan to work more in the down years, rather than rely on a cash/bond bucket. Since I’m in my 30s and plan to maintain my skills during early/semi retirement (I’m a professional software developer), I plan on picking up some consulting work in the particularly bad years. Not only will this allow me to keep most of my portfolio in equities and provide additional capital to invest when prices are low, it will also make me appreciate my early retirement when I go back to real work for a few months 🙂

    I understand that it is usually difficult to find work during downturns but based on the current state of the software industry and the fact that my job can be done from anywhere (which means I could potentially work for anyone), I feel confident that I would be able to pick up additional work when necessary.

    I hope you and Caroline have a great Christmas and I look forward to reading more about your retirement income strategies in 2014!

    • Thanks Mad Fientist. Your strategy makes sense, and you are fortunate to have one of those select career skills that can sell even in a downturn. Thanks for the holiday wishes, and stay in touch!

  2. Darrow, it’s nice to see that people still remember Bob Clyatt’s 4%/95% guideline. That was a big (and expensive) research project when he published his “Work Less, Live More”. I’m hoping to see that type of logic become standard in more variable spending simulations.

    I’ve learned that keeping two years of expenses in cash (6%-10% of a portfolio) can help a lot during a bear market. We replenish the cash stash when the markets are up. Spending it down for two years will get you through almost any recession. Wade Pfau has looked at this a little bit but there’s lots left to be done.

    • Thanks Doug. I appreciate Clyatt’s guideline. It’s a simple way to capture the idea of cutting back in a downturn. Some of the better retirement calculators offer similar logic. And thanks for your real-world experience. Good to know that two years of cash has bridged most recent recessions. I’m interested in any research in this area too.

  3. derek castle says:

    thanks again for a helpful article Darrow. To be little picky (probably quite unnecessarily) the quotation you gave from Wade Pfau valuations article doesnt, I think, give the right impression of what he was trying to say. A better quote would have been “Efforts to elevate market valuations into an overwhelmingly central role that trumps all other considerations are not well-founded”. I believe Wade does see valuations as an important consideration, just not the “be all and end all” of considerations: there are other considerations as well. You may not have meant it,but the quote you used made it sound like Wade was arguing against those who advocate against having lower withdrawals when markets are high. But, as I am sure you know, Wade does view valuations to be an important consideration: just not the only one.
    But , sorry ,that was probably way too pedantic : your article was a great summary. Thank you

    • Thanks for the clarification Derek. And thanks for finding that quote, which adds some helpful perspective. I agree. My single paragraph probably can’t do the topic justice. Wade has written quite a bit on the subject. If somebody wants a complete understanding of his views, I’d recommend his articles.

  4. Excellent review. Thanks so much for this.

  5. Ralph Parker says:

    Thanks Darrow for this article.

    I’ve been reading Morningstar’s bucket system but you said one system was discredited. Could you quickly summarize it? Morningstar has stress tested their bucket approach as I’ve seen in their articles a few months ago.

    I’ve been working on my retirement spending plan and I’ve incorporated a bucket approach that tries to keep money I need in a 5 year window outside the stock market. What I like about the bucket system is that it lets me see how to optimize my allocation and contain my risk. You always replenish your cash from stocks in an up market but in a severe down market you get your cash from the conservative buckets. The catch in this is that bond funds appear scary now since the interest rate can’t go lower and are more than likely to rise leaves me the impression that bond funds are a bad gamble in these times.

    Now couple the bucket system with a withdrawal plan that I will call 99% Probability of Success (POS). The annual spending rate is based upon a montecarlo like flexible retirement planner that allows you to input actual streams of income (pension, short term pension supplement, wife’s pension, two Social Security streams) an planned expenses (like buying a car or purchasing a home) It statistically models the markets so you can input a spend rate that gives you a 99% POS. Each year or in the event of a significant down market you recalculate your 99% POS rate and adjust your allowance. The 99% POS gets adjusted each year based upon life expectancy changes and market growth. In my studies I did not let the spend rate grow faster than inflation. I believe that changes made the earliest possible when the market turns down provides more salvation to the spend plan that delayed changes. In my modeling of my own financial scenario being flexible like this turns your spend plan into a lifetime safe plan however, 97% POS did not work as it produced financial failures (totally running out of money/savings).

    • Hi Ralph, thanks for the details on your spending plan. Sounds like you have done a very thorough and careful analysis for your personal income and expenses, which is always going to be most accurate.

      If you can afford a 99% POS, my hat is off to you. The downside of that high margin of safety might be that you leave lifestyle on the table in many possible outcomes. Some experts suggest lower probabilities of success are OK.

      Your bucket implementation sounds similar to mine, in general. As you point out, in today’s climate, we can’t put all our faith in bond buckets.

      It’s the “Buckets of Money” strategy that came under fire: you can Google for details.

      Thanks again for the detailed comment.

      • Ralph Parker says:

        Thanks for your response,

        For me the bucket system is a little more than a tool to set allocations. Mainly, It allows me to decide how much money to put in each risk category. To do the bucket system: I have to determine a spend rate program; I have to evaluate investments based on income and risk; I have to develop a risk mitigation plan (which I’m still working on) that helps me work through the market downswings; and then on great markets I have to have a give me a raise plan. After all I don’t want to leave too much money to my heirs. The ‘Buckets of Money’ approach as defined in the Bogleheads column did seem awkward and somewhat contrary to my instinct as described earlier.

  6. Nicely done Darrow!

    I just made a link to this post an Addendum here:

    Which is turn is a guest post from the Mad Fientist who commented above. Circle of life. 😉

  7. Jerry Dion says:

    I am new to your site, as my wife and I are coming up to late-early financial independence in early 2015. We’ve run several of the retirement models (Quicken, T Rowe Price Future Path, and Market Watch Visual Planner) that all show high confidence in the assets-income-spending scenarios out to age 95. Establishing the specifics of a withdrawal strategy is our next task and this post is a good tour of the many ways to go about it. Right now we are thinking about delaying Social Security through an optimizing approach and using my modest inflation adjusting pension and our retirement accounts to fund the first six to eight years of financial freedom. The way I look at this is that we are essentially “buying” increased inflation adjusting SS annuities by using our asset accounts, and should be able to cover 80% of planned spending with pension and SS when we both hit 70 (my wife is a couple years younger). We’ve been steady dollar cost averaging investors, aggressive on the equities allocation in index funds, ignoring all market swings (despite some queasiness at times). But how to manage taking money out in a tax efficient and sustainable way is the tricky part. I’m hoping your site can help in getting our heads around the ins and outs of this.

    • Hi Jerry, welcome, and thanks for the comment. Congrats on your planning. You’ve certainly prepared well and done all the right things leading up to retirement. I agree with your views on Social Security. Everything I read tells me it makes sense for those with the means to delay taking it as long as possible. (Though in some cases it may be a good compromise to have the lower-earning spouse start sooner.) FYI, my go-to reference for dealing with Social Security is Mike Piper’s excellent book: Social Security Made Simple.

      Withdrawing from retirement assets in a sustainable and tax efficient way is going to be a main focus of mine. I’ll be learning as I go, and sharing the details as best I can. I’m always interested in others’ insights too, so feel free to add yours. Thanks again.

      • Jerry Dion says:

        The SS optimizing strategy includes my wife making a restricted application for spousal benefit after I apply for SS at FRA and then suspend my benefit until 70 1/2; she would drop the spousal benefit and apply for her own at 70 1/2. We both would receive close to maximum SS benefit amounts.

  8. Well written as always. This is a key topic for retirees and I will be looking forward to your future posts. I wish you and Caroline a happy holiday season.

  9. DK-

    Great article as always. We are in the edge of RE and plan to use a VWR similar to Clyatt’s method, which you mentioned. We will use the Guyton-Klinger approach, which is a kind of “guard rails” system. For those considering a VWR approach, G-K is worth a look. You can of course find it with Google but, can also find threads discussing it on

    Happy Holidays!

  10. Always interesting writing, thanks Darrow.

    I still think the annuity portion gets a bad rap, depending on which annuity you buy. I have had mine with Vanguard now for over a year and I couldn’t be happier. I can still take all my money out tomorrow with no penalty if I want, they just bumped up my monthly payment (because the Wellington mutual fund the entire annuity is in has increased in value over the previous year) and, if the market drops by 50% (God I hope not), I will still get the larger monthly payment.

    I do agree, I would never hand over my money to an insurance company and kiss it good-bye….

    Mele Kalikimaka!!!