Is 2 Percent the New Retirement Safe Withdrawal Rate?

safe withdrawal rateWhen planning for retirement, it’s not just about how much you money you save up. One of the most important variables to consider is what’s called the safe retirement withdrawal rate – how much money you’ll withdraw each year to live off of. To take too much may drain your nest egg and leave you with nothing. To take too little may curb your lifestyle more than you’re use to.

For years, financial advisers have long proclaimed a safe withdrawal rate of 4 percent from your retirement accounts as the rule of thumb.  That means you would have a relatively high chance of success if you start off by taking out only 4 percent and then adjust each year with the rate of inflation.  But lately not everyone agrees, and the spread is more surprising than you think.

A New Spin on the Retirement Safe Withdrawal Rate:

Smart Money magazine recently ran a great article in their February 2012 “New Retirement” section by Glenn Ruffenach that investigated the viability of the 4 percent rule. In the article, he cited:

• The Journal of Financial Planning predicted that a retirement safe withdrawal rate from nest egg is 1.8 percent.

• The Retirement Management Journal said some individuals may be able safe withdrawing as much as 7 percent during retirement.

So if the experts can’t agree, what are you supposed to believe?


Where Did the 4 Percent Rule Come From?

The 4 percent rule as a retirement safe withdrawal rate was developed in the 1990’s by a certified financial planner from California named William Bengen. It was meant to establish a high probability that your money would last for the next 30 years. Bengen’s findings became a popular rule of thumb for planners and analysts everywhere to use in helping their clients with retirement planning and professional tax software.

My Thoughts:

So what do you believe will make a safe withdrawal rate?

For me, 7% sounds pretty wild! I don’t think I’d chance my life savings with a withdrawal rate that high.  Yet, 2% sounds a little too conservative.

When I work on my own money design, I prefer to keep things easy. Here is the formula I use for planning my retirement:

• My balance with grow by an average of 6% annually.

• Inflation will decrease my portfolio by an average of 3% annually.

• Therefore, I should be able to safely withdraw an average of 3% annually from my retirement balance.

Sounds simple, right? If your money grows by 6% but decreases by 3% for withdrawals and another 3% for inflation, then your money should (in theory) last for virtually the rest of your life. I purposely use 6% instead of 8% for predicting future returns as a safety buffer against the unforeseen.  In addition, these figures will also help foster a consistent withdrawal rate. In other words, you shouldn’t start off taking out a large sum of money only to be left taking out lesser amounts of money in the years to come as your nest egg disappears.

The following graph is an illustration of what a portfolio of $1 million dollars would look like for the next 50 years depending on which number believe is a safe retirement withdrawal rate. You can download my Excel worksheet and try your own figures yourself.

safe withdrawal rate, Nest egg, withdrawal rate, Smart Money, 4% rule, Journal of Financial Planning, Retirement Management Journal, William Bengen, financial planning

Download My File: MyMoneyDesign_Retirement_Withdrawal_Example.xls

As you can see, the higher the safe withdrawal rate you pick, the more likely you are to run out of money.  My preference for the 3% figure puts me as close as possible to nearly “breaking even” each year.

Testing Your Retirement Safe Withdrawal Rate:

I will admit there is a problem with my model as well as all “rule of thumb” calculations:

• Portfolios don’t necessarily grow by a certain percentage every year. In fact, some years they decrease.

Failure to consider this truth is why sometimes our models do not turn out the way we think they will.  To really “test” our safe retirement withdrawal rate, we’ll need to create a simulation of the next 30 to 50 years.  You can try this out for yourself in my next post: A Better Way to See If You’ll Run Out of Money During Retirement.


Related Posts:

1) Are We Fools for Saving Our Money?

2) A Strategy for Maxing Out Your Retirement Savings

3) Why Do You Want to Retire?

Photo Credit: Microsoft Clip Art


  1. says

    3% real return is far below historical averages, but I personally prefer 3 to 3.5% myself. 4% does have a record of going broke at some points in the past. Also, you can always adjust. During the good years you can go down to 2% and less.

    • MMD says

      We are in agreement. 4% seems a little too high for me. I need my money to last for longer than 30 years. Plus I think the game has changed a little since the 1990’s. The 3% range is where I feel comfortable. If I get too far into old age and have too much money left over, I’m sure there will be people there to help me spend it.

      • says

        There will always be people there to help you spend it! haha We’re retired with a decent pension. So we don’t withdraw anything from our nest egg and plan to keep it that way as long as possible. Why? 1) It’s not as big as we’d hoped due to market downturns. 2) We are satisfied with our current modest lifestyle and don’t need to tap the egg.

        • MMD says

          Thanks Maggie! That’s the key right there – being happy with your modest lifestyle. As long as your wants are under control, your money should last you a long time. I too hope to hold out for as long as possible on making my withdraws.

  2. BusyExecutiveMoneyBlog says

    This is an awesome post on a topic that I plan to spend a lot of time on over the next couple of years. We all want to build our passive income streams but some level of “nest egg” is hard to completely forget about. Given that I plan to retire early, i think 3% makes more sense.

    • MMD says

      Thanks for the compliment! I’m with you. I plan to retire early as well, and therefore I’m aiming way lower than all conventional figures. I’ve also spent a lot of time building models and then deciding they were way too risky. The only conclusion I ever come to is that defense is the best strategy against never running out of money. I think you’re really going to like the next part of this post when I have a Monte Carlo simulation for you to download and play with.

  3. says

    I am big fan of velocity over time vs. static picture. You did a great job in briefly showing this in this post. I also like how you provided basic rules of thumb upfront with nice metaphors. Like it. I will run my numbers and see what 3% means to my style and how it reflects on the velocity.

    • MMD says

      I’m glad to hear someone downloading these spreadsheets and putting them to use! I’ll have a more realistic (but more complex) model for you to try in the next post.

    • MMD says

      Thanks! I’m glad to hear you like my logic. I actually didn’t know how old that rule was or the origins until I read it in Smart Money. The 90’s doesn’t seem that long ago to me. But I could easily see how recent events would cause people to second guess these traditionally accepted values.

  4. says

    I’m risking getting shot here (good thing nobody can actually see me face-to-face), because these “products” have gotten a bad name over the past 15 years.

    I’m with you on the withdrawal rate and being conservative (ESPECIALLY IF YOU’RE PORTFOLIO IS AT RISK IN THE MARKET), but there are annuities out there that are paying 6.5% guaranteed interest (as long as you agree to take your money back in monthly/yearly payments over the course of your life).

    Depending on your age, the payouts range from 3.5% – 5.5% of your asset. If you’re in your 50’s or 60’s it’s likely they’ll pay 4.5 – 5% withdrawal rate.

    So, if you’re doing your own investments and trying to draw a lifetime paycheck, then yeah, I’d be more conservative. However, if I have an insurance contract that guarantees me 4.5-5% then I might go that route too. Good thing I have years before that happens. :)

    • MMD says

      Don’t worry – no shooting here! I will be perfectly honest that I intend to look further into annuities. I really only know a minimal amount about them. I know there is a lot of negative press about them. Yet, annuity sales seem to increase every time I read about them. So there is something about them worth investigating.

      4.5 to 5% sounds incredible. But is that figure net or gross? What about fees? I do know annuities are notoriously expensive whereas mutual funds can be anywhere between 0.2 and 1%. What about taxes? Is the income taxed at the same rate? And then there’s the big problem with the fact that your money goes to the insurance company rather than your heirs after you die. Assuming there is anything left of my nest egg after I pass away, I’d like to keep it in the family.

      Like I said, I’m not knocking annuities because I do believe there is a place for them. These are just a few questions I have right off the bat based on what I know about traditional investing.

      • says

        Those are all great questions!

        Just like most things, there isn’t 1 “type” of annuity. There are fixed annuities, variable annuities, and fixed indexed annuities.

        Variable annuities are generally the ones with the ridiculous fees: you get the .2-.5% fee for your money being invested in mutual funds and then you pay another 2-3% in other fees.

        Frankly, the ones I like the most are fixed indexed annuities.

        About the taxability: annuities are simply insurance contracts and are primarily set up via an IRA. So IRA tax rules apply. If you have a Roth IRA then there is no tax, if you have an IRA then you pay standard income taxes as they money is paid to you. If you have non-qualified money, then the gains are taxable but your contributions are not.

        The 4.5-5% is based on the gross number and you’re taxed based on the above scenarios.

        For FIAs (Fixed Indexed Annuities) there typically aren’t any fees for the underlying insurance product. However, there is an INCOME Rider (roughly .6% fee) that does the guarantees that I mentioned: that 6.5% interest rate and the paycheck for the rest of your life.

        It is complicated though and that’s why they get a bad rap: people don’t fully explain them (because many don’t know what they’re doing) and a lot of people get “sold” a product that they don’t fully understand how it works.

        The problem with annuities is liquidity. A part of the tradeoff if that your money is tied up…it’s very much like a pension. In fact it really is a pension except there still may be a “death” benefit based on the amount of money leftover.

        • MMD says

          Jason: Thanks for doing all my research for me! :) I have can see I have a lot to learn about them; especially the aspect of the fees. This is where people really seem to disagree with these products. However, that doesn’t scare me. I’m sure there are good ones out there.

          Without getting ahead of myself, I’m strongly considering a strategy for retirement where I take a portion of my nest egg (say 10%) and spend it on an annuity. The prospect of lifetime guaranteed income would serve as a great hedge against economic troubles.

          One other question for my resident expert: Do the interest rates on these products change with the national rates? Right now, interest rates are at all time lows. When they go back up someday, could it be possible to lock into one at perhaps 8 to 10%?

          • says

            Annuities are paying you earnings and capital. You get any sort of life annuity and it stops at death. Some have a guarantee for a term. For example you can get a life annuity guaranteed for 10 years, which means that someone gets a payment for the first 10 years. If you died before the 10 years are up, your heirs get the money for the 10 year period. You will get it for life.

            Annuities can be great products, but be sure you know what you are buying and what the risks to you are.

            • MMD says

              Thanks for the answers! I definitely have some work to do on researching these products. I totally believe there is a place for annuities. But like everything else, you’ve got to educate yourself so you know the right questions to ask ahead of time.

  5. says

    I’m glad to see you writing about this, MMD.

    This is Rob Bennett. I am the person who discovered the errors in the Old School safe withdrawal rate studies. I reported on the errors in a May 13, 2002, post to a Motley Fool board and a group of us got together and did analytically valid research that reports the real SWR. I have a New School calcualtor (“The Retirement RIsk Evaluator”) at my site and lots of podcasts and articles and other materials on this issue.

    The big mistake here is trying to come up with a single number. People say “well, 7 percent sounds too high” or “2 percent sounds too low” or whatever. When you take into consideration your feelings about what the number should be, you are engaging in rationalization, not science. We all were influenced by the insane bull market. Rather than let our feelings influence our take on the numbers, we need to let the numbers rein n our emotions. The numbers did not get caught up in the craziness of the bull. The numbers are objective and thus more trustworthy than out subjective emotions.

    The objective reality is that there are times when the SWR is 1.6 percent and there are other times when the SWR is 9 percent. And it is very easy to see why both of those numbers are perfectly appropriate and reasonable numbers.

    At the top of the bull, stocks were priced at three times fair value. In 1982, the bottom of the bear that preceded the huge bull, stocks were priced at one-half of fair value. So stock valuations increased by 600 percent over that time-period (from 0.5x to 3x). So we should expect the SWR to DROP by 600 percent over that time, no?

    When you use an analytically valid methodology (one including an adjustment for valuations), that’s just what it does. The Risk Evaluator shows that the SWR for retirement beginning in 1982 was 9 percent. It shows that the SWR for retirements beginning in 2000 was 1.6 percent. Round it down to 1.5 to make the math easy and multiply by 6 and you get 9 percent. The data confirms that stocks perform just as common sense says they should perform!

    The reason why people are confused about this is that Wall Street has spent hundreds of millions of dollars of marketing money trying to persuade us all that Buy-and-Hold strategies can work. There is zero support in the academic research for this claim. Valuations have always affected long-term returns. So stocks are more risky when they are high-priced and investors trying to keep their risk profiles roughly constant need to be certain to adjust their stock allocations in response to big price shifts.

    That’s market timing! Short-term timing really doesn’t work. But long-term timing is absolutely required for investors seeking to have a realistic chance of achieving long-term success. If you are not changing your stock allocation in response to big price shifts, you are letting your risk profile get wildly out of whack.

    Again, thanks for the super article. We need more like this!


    • MMD says

      Rob: Thanks for this very well crafted comment! You are in good company that I agree we should let data guide us rather than use our emotions. Although I follow your logic about the safe withdrawal rate being anywhere from 1.6 to 9 based on how stocks are valued and market conditions, I think most people will still try to search for that one single number. Evaluating market valuations is a complex approach for the common do-it-yourself investor. On top of that, it is difficult to know whether or not the market is under or overvalued until we see what the closing prices are and it becomes past history. I would assume in that regard your withdraw would need to be based on the data from the year before. Could it not be argued that to use a lower withdrawal rate would provide a hedge against market conditions? For example, if we are in a period where 9% may be appropriate and I still opt for 3%, then I have only continued to save my money for a later date.
      Again, thanks for adding this valuable comment.

    • MMD says

      “… Wall Street has spent hundreds of millions of dollars of marketing money trying to persuade us all that Buy-and-Hold strategies can work. There is zero support in the academic research for this claim.”

      After crunching the numbers on The Lost Decade, I am not totally convinced that buy and hold is really the best alternative either for our present day situation. But what puzzles me is how come an Index Fund works if buy and hold doesn’t. For example, in my Excel file for the next post, I have calculated that the S&P 500 has returned an average annualized 8% since the 1950’s. So in reality if I had bought all those stocks years ago and held onto them through thick and thin, I’d have an average of 8% every year. Isn’t this evidence of buy and hold working?

  6. says

    Could it not be argued that to use a lower withdrawal rate would provide a hedge against market conditions?

    It provides a hedge. But at a big cost.

    If the true SWR is 8 percent and you use 4 percent, you indeed have a lot of slack in your plan. But you will need twice as many assets to be able to retire! That might mean 2 million when 1 million would have done the job. I question whether investing analysis means much when we are permitting ourselves to be that far off the mark in our calculations.

    You are right that the average person isn’t going to go to the trouble to calculate valuations without encouragement. I believe the experts need to make it a practice always to incorporate valuations into their analyses. Then it just becomes part of the everyday discussion of how stock investing works.

    For example, when the radio reports that “the DOW is at x,” they should add a comment that “stocks are today priced at three times fair value, so the true value of your portfolio is one-third of the number you see on the bottom of your statement.” That one change would change stock investing in a dramatic way. We would never again have another bull market. Which means that we would never again have another bear market. Which means in all likelihood we would never have another economic crisis.

    Valuations are easy to overlook and for a time (when the extent of the overvaluation is small), they really do not matter much. But it’s like gaining weight. It’s not a big deal to be 5 pounds overweight. But if your response to the problem is to avoid getting on the scale anymore, you are soon going to be 10 pounds over and then 20 pounds over and then 40 pounds over. Then you have a heart attack and wish you had paid attention to the problem when it would have been easy to solve it.

    We investors let our weight problem get totally out of hand in the late 1990s. That’s why we are in an economic crisis today. All that is really happening is that we are losing all the fluff we took on during days when we ignored the problem of overvaluations and the loss of those trillions upon trillions of pretend wealth has put us in a state of shock and caused millions of us to become afraid to spend and thereby support economic growth.

    Thanks again, MMD. Please take care.


    • MMD says

      If the SWR is always changing based on market valuations, doesn’t that also create intense variations in how much I’d need to have to retire or how much income I’d receive? For example, let’s say I use your numbers and save $1M. At the time of retirement, the market allows for an 8% SWR so I withdraw $80K. But a few years later, the SWR is 4%, so now I can only withdraw half ($40K). This would cause huge problems for my retirement cash flow. Perhaps I really needed $2M to do the job. But now it is too late because I have already retired.

      Instead, could I not use an “average” of the SWR and base by nest egg target on this amount?

      • says

        If the SWR is always changing based on market valuations, doesn’t that also create intense variations in how much I’d need to have to retire or how much income I’d receive?

        Super question, MMD.

        If you use the correct SWR on the day you retire, that will remain the correct SWR for you for the entire length of your retirement. But a very different SWR could apply for a retirement that begins a few years later.

        Say that stocks are priced at fair value and that the SWR is 4 percent and that you retire with $1 million. So you are able to safely withdrawal $40,000 (inflation adjusted) for every year of your retirement.

        In Year 5 of that retirement, valuations double and the SWR drops to 2 percent. Your SWR remains 4 percent because the 4 percent is being applied to your initial portfolio value of $1 million; that number doesn’t change as a result of the price run-up. If the fellow retiring in Year 5 also had $1 million saved in Year 1, his SWR is 2 percent. But it is 2 percent of $2 million (his Year 5 portfolio value), not 2 percent of $1 million. So he is taking out the same amount as you.

        The root problem here is that we all should be discounting our portfolio values for the effect of overvaluation. A $100,000 portfolio value when we are at the valuation levels that applied in 1982 (one-half of fair value) has the same lasting real value as a $600,000 portfolio value when we are at the valuation levels that applied in 2000 (three times fair value). All that matters is the real value of a portfolio. Both overvaluation and undervaluation are temporary phenomena. John Bogle has pointed out that Reversion to the Mean is an “Iron Law” of stock investing.

        I understand that the claim being put forward here is very bold indeed. I have studied this matter on a daily basis for 10 years now and I can assure you that there is 30 years of academic research backing up the claim.

        We didn’t know everything there was to know about how stock investing works back in the early 1970s, when the Buy-and-Hold Model was being developed. Had Shiller’s 1981 research been published in 1971, the book would have been titled “A Valuation-Informed Walk Down Wall Street” and we all would be Valuation-Informed Indexers today. Because we didn’t have all the research we needed available to us at the time, we got on the wrong track with some things and we are today in the process of beginning to sort out all the confusion.

        Please don’t take my word for it, MMD. I wouldn’t want that. You need to check out the research. If you (or others) are interested in following up, I can point you to lots of materials, including research by tenured professors. This is very exciting stuff. It changes everything we thought we knew about how stock investing works in very exciting and positive ways. When we know how valuations work, we can predict long-term returns, and when we can predict long-term returns, we can dramatically minimize the risk of stock investing.

        Anyway, thanks much for your question. It gets right to the heart of things. I’ll respond to the second part of your question in a follow-up comment.


      • says

        Instead, could I not use an “average” of the SWR and base by nest egg target on this amount?

        This is precisely what those using the Old School SWR studies as their guide are doing, MMD.

        The 4 percent number is roughly the number you would get if you took the SWRs that apply for all the valuation levels we have seen in U.S. history and averaged them. The highest SWR we have seen was the 9 percent SWR that applied for those who retired in 1982. The lowest SWR we have seen was the 1.6 SWR that applied for those who retired in 2000.

        If you happened to retire at a time of moderate valuations, the 4 percent number would work well. If you happened to retire at a time of super low valuations, using the 4 percent number as your guide would unnecessarily delay your retirement by many years. A retirement beginning in 1982 that called for a 9 percent withdrawal stood a 95 percent chance of working out. That’s pretty darn safe. Someone planning to retire who needed $90,000 to live on and who had a $1 million portfolio was set. If he used the 4 percent number, he would have been fooled into believing that he was only halfway to his goal and would need to work decades longer before he could retire safely.

        There are millions of people who used the 4 percent number to plan retirements during the bull years. Almost all of those people will be seeing their retirements fail in days to come, in the event that stocks continue to perform in the future anything at all as they always have performed in the past. The historical data shows that those who retired in 2000 using a 4 percent withdrawal have a one in three chance of seeing their retirements survive for 30 years.

        The millions of failed retirements we are going to see as a result of our use of the average SWR rather than the valuation-adjusted SWR is going to be one of the biggest social crises we have ever had to endure as a nation. It is going to cause bigger political problems that those caused by the Great Depression. It is going to require more costly government bailouts than those that were adopted in the early days of this economic crisis.

        I believe strongly that we should be using accurate numbers. I know from experience that many people object strongly to this. I am unable to explain the hostility to the idea in any way other than to say that people don’t feel comfortable with new ideas until they have been widely discussed or people are emotionally invested today in Buy-and-Hold. Taking valuations into consideration is a win/win/win. It’s not just that it provides us access to accurate retirement numbers. It permits us to reduce the risk of stock investing by 80 percent.

        Again, that’s a bold claim. But, again, the entire historical record backs it up. We have data going back to 1870. In those 140 years, there has never been a time when buying stocks at a time of moderate or low prices did not generate great long-term results. There has also never been a time when buying stocks at a time of insanely high prices generated anything less than a wipeout of most of the investor’s accumulated wealth of a lifetime. Given these realities, what rational case can be made for sticking with the same stock allocation at all times?

        This is a paradigm shift, MMD. That’s why many people have a hard time taking to it. Thinking about things in these terms changes everything we have ever believed about how stock investing works. But I wish that people could see that the changes are all for the good. If this is really how stock investing works (and, again, there is now 30 years of academic research showing this to be so), we can take most of the risk out of stock investing by exploring this path. And our returns would increase! High returns with low risk — That’s investor heaven!

        We’re almost there. I believe strongly that this is the next step. We just need to get these ideas before more people so that we can explore them from every possible angle before adopting them. We need to be sure before going ahead. But everything I have seen over the past 10 years tells me that, the more we explore and debate these ideas, the more we will grow in confidence that these are the ideas that truly work for the long-term investor.


    • MMD says

      “For example, when the radio reports that “the DOW is at x,” they should add a comment that “stocks are today priced at three times fair value ..”

      If the media reported what stocks were priced relative to fair value, would we even have a need for analysts? – Just kidding!

      I see your point. But in reality, don’t you think there would a lot of disagreement about what the fair market value is?

  7. says

    If the media reported what stocks were priced relative to fair value, would we even have a need for analysts? – Just kidding!

    Please understand that what I am describing only works for index funds. It is Bogle’s introduction of index funds that started what I believe is an ongoing revolution in our understanding of what works for middle-class investors.

    No one can say in advance how an individual stock will perform in the long run. The company might release great products, like Apple, or the company might go belly up. So there are always going to be significant risks attached to investing in individual stocks.

    Long-term returns of broad index funds are highly predictable using valuations, however. The reason is that, when you buy the entire market, you get a mix of the good companies and the bad companies. You don’t need to identify which companies belong in which category. Your return is determined by the productivity of the U.S. economy as a whole. U.S. productivity has been sufficiently strong to support an average long-term stock return of 6.5 percent real for as far back as we have records. That number could change a little in either direction on a going-forward basis. But it is not likely to change dramatically.

    The problem many of us are having today is that we are not adjusting our stock allocations in response to big changes in valuations. It’s obviously not realistic to expect stocks to offer the same long-term value proposition when they are priced at three times fair value as when they are priced at one-half fair value. The data shows that the most likely annualized 10-year return in 1982 was 15 percent real. The most likely annualized 10-year return in 2000 was a negative 1 percent real. What one stock allocation makes sense in both circumstances?

    There would still be a need for analysts to look at individual stocks. But the indexer (I believe that most middle-class investors should be investing primarily in indexes) would not need to worry about any of that. The idea here is to provide the typical middle-class person with a simple, safe strategy that also provides returns plenty high enough to finance a solid middle-class retirement.


  8. says

    But in reality, don’t you think there would a lot of disagreement about what the fair market value is?

    It doesn’t solve the problem for us all to pretend that the nominal value of our portfolios has some special significance. Stocks were priced at three times fair value in 2000. Say that you were 60 years old at the time, had a portfolio with a nominal value of $900,000, and know that you need $1 million at age 65 to be able to retire. You’re thinking you are set. Even if you make no further contributions, growth on the amount already invested is going to get you where you need to be in five years.

    Now consider the reality. You have a portfolio with a real, lasting value of $300,000. You are cooked. You have just about no chance of being able to retire at age 65. Shouldn’t we be telling that person where he stands instead of indulging his fantasy beliefs that he is on the right track?

    That fellow has to make decisions every day about what to do with his money. He might be thinking of adding an addition to the house. He might be thinking of buying a second or third car. He might be debating between a vacation at the local beach or a cruise of the Caribbean. How can he make good decisions if he has no idea of how much money he really has in his portfolio?

    Valuations is 80 percent of what effective stock investing is about. Leave valuations out of the analysis and you cannot get anything else right.

    Would there be arguments about the right valuation metrics to use?


    There is a body of research showing that P/E10 (Shiller’s metric) is best. I think that most responsible people would go with that. But a case could probably be made for some alternatives (Tobin’s Q has a good track record also).

    There is no harm in differences of opinion. If some want to use P/E10 and some want to use Tobin’s Q, what’s the harm? They will debate their respective positions and we can all listen in and decide for ourselves which case is stronger. That’s the way we decide things in every area of life pursuit outside of stock investing. Why wouldn’t it work with stock investing too?

    The dangerous thing is to avoid the disagreements by pretending that the issue doesn’t matter. When the media reports non-adjusted numbers, they plant the idea in people’s heads that the numbers are real. They aren’t. People get hurt when they make financial plans based on numbers they perceive as being real that in fact are not even close to being accurate.

    Let the disagreements begin! That’s what I say. Let a thousand flowers bloom!

    At least if people hear that there are disagreements, they come to understand that this is an issue they need to look at. The lack of open disagreement has led many to believe that the non-adjusted nominal portfolio number is the neutral choice. Is it? I sure don’t think so. There are circumstances in which using the non-adjusted number is an extremely dangerous practice.

    In January 2000, the most likely 10-year annualized return for stocks was a negative 1 percent. The certain return on Treasury Inflation-Protecged Securities (TIPS), a risk-free asset class, was 4 percent real. That’s a 5 percent differential. Not for one year. For every year of 10 years running. Do the math and you come to a total differential of 50 percent of the initial portfolio value.

    For someone with a portfolio in 2000 of $100,000, that’s $50,000 lost because as a society we have elected to avoid discussions/arguments about valuations. For someone with a portfolio of $500,000, it’s a $250,000 loss. For someone with a portfolio of $1 million ( we all need to get close to $1 million before we can retire), it’s a $500,000 loss.

    Even those numbers do not reflect the total loss associated with our collective decision to avoid discussions/arguments about valuations. We all know about the power of compounding returns. If you suffer that $250,000 loss at age 50, you will be losing out on the compounding you otherwise would have enjoyed on that $250,000 for the next 30 years (presuming that you die at age 80).

    It is because we are all in the process of suffering losses of this size that we are in an economic crisis today, MMD. If you look at the historical record, you will see that we have never once in U.S. history since 1870 (that’s as far back as we have records) suffered an economic crisis without first going to a P/E10 value of 25. It works the other way around too. Never have we gone to a P/E10 value of 25 and not seen an economic crisis. The worst economic crisis in history was the one caused by the highest P/E10 value — We went to 33 in the months before the Great Depression. We went to 44 in early 2000.

    Economic crises are caused by the shock that investors feel when they see themselves losing the accumulated wealth of a lifetime. This causes people to pull back on spending. When millions pull back on spending, tens of thousands of companies fail. When tens of thousands of companies fail, millions of workers lose their jobs. Political unrest often follows.

    All of this is unnecessary. If people understood that the long-term value proposition of stocks drops as prices rise, people would sell stocks in response to big valuation jumps. That would bring prices back down. Problem solved! Market prices are self-regulating as long as we are sure always to let people know about the effect of valuations on long-term returns. It is when we stop talking about this stuff that the mad cycle that causes economic depressions gets started.

    We don’t all need to agree on every aspect of the valuations question. We only need to all agree that we should all be free to talk about it. It is the discussion that informs us of the realities. Shutting off the discussion puts us in the position of the overweight person who decides that the answer to his problem is to stop getting on the scales. Not knowing permits us to get carried away with bull markets that can do us all great damage a few years down the road.

    That’s my take, in any event. Lots of good and smart people think I am out to lunch re all this stuff.


  9. says

    So in reality if I had bought all those stocks years ago and held onto them through thick and thin, I’d have an average of 8% every year. Isn’t this evidence of buy and hold working?

    This is another super question! I love your questions, MMD. This one gets to the question of why there is so much confusion re how valuations affect returns.

    I have a calculator at my site called “The Stock-Return Predictor.” The purpose is to help people set the stock allocations they should be going with at different valuation levels. The calculator performs a regression analysis on the historical data to reveal the most likely annualized 10-year return starting from any valuation level.

    We cannot predict long-term returns precisely. What we can do is to identify a range of possibilities and assign rough probabilities to each point in the range.

    In 2000, the most likely annualized 10-year return was a negative 1 percent real. This is obviously very bad given that the average log-term return is 6.5 percent real. But the calculator also reveals the most likely 30-year return. January 2000 was the worst time in the history of the U.S. market to be buying stocks. But do you know what the most likely 30-year return was? 5 percent real!

    That’s not too bad at all. It’s certainly better than what you can usually get in super-safe asset classes like TIPS and IBonds and CDs.

    So there is a sense in which it really is true that stocks are always best for the long run. At the conclusion of 30 years, stocks will always do best.

    But, when stock valuations are where they have been since 1996, stocks get beat over 10-year periods by just about everything else out there, including money markets. Even money markets don’t pay a negative return.

    Buying an index fund is like buying a percentage of the growth in the U.S. economy. You’re buying a piece of paper that entitles you to a 6 percent real return per year. That’s an amazing deal. Stocks are a great asset class.

    But the price you pay matters. If you pay a fair price for that piece of paper, you really do obtain the 6 percent return. If you pay double fair price, you only obtain 3 percent (half of your money is going to the purchase of cotton-candy nothingness). Pay triple fair price and you only get 2 percent. Pay half of fair price and you obtain 12 percent (every dollar you pay gets you two dollars worth of that paper providing you a return of 6 percent real). The price you pay makes a HUGE difference.

    BUT —

    After the passage of 30 years, you are going to have experienced both a secular bull and a secular bear. At that point, the times at which stocks pay 12 percent have been balanced with the times at which stocks pay 2 percent and your return is indeed likely to be something in the neighborhood of 6 percent real.

    In the short term (time-periods of less than 5 years), stock prices are entirely unpredictable. They are set by investor emotion and no one has figured out a way to determine in advance in which direction investor emotion is headed. Short-term timing NEVER works. The Buy-and-Holders are right about this one and we should all be grateful that they brought this powerful insight to our attention.

    In the super long-term, the good and bad years even out and stocks always provide a good return. Stocks really always are best for the super long-term.

    It is in the time periods from 10 years out through 20 years out that valuations are the key factor. Today’s valuation level does a very good (but not perfect or precise) job of predicting the 10-year return. It does an even better job of predicting the 15-year return. And it does an outstanding job of predicting the 20-year return.

    The trouble with not looking at valuations is that it sets you up for such damaging losses at the turning points (which cannot be identified in advance). We were at three times fair value in 2000. We always go to one-half of fair value in the bear market that follows a bull. This means that those who were heavily invested in stocks in 2000 will be losing five-sixths of their accumulated wealth of a lifetime before we reach the end of this bear market. Someone with $600,000 in 2000 will have roughly $100,000 remaining at the point we reach a turnaround.

    Then we will see 10 years of 15 percent annual returns!

    The guy who went from $600,000 to $100,000 will make it all back and then earn a 6 percent real return on top of that by the end of 30 years. But look at the hell he has to live through to get there! Can any middle-class person realistically be expected to stick with a high stock allocation through those sorts of losses? Most are forced to sell at some point and all bets are off for that group.

    You don’t have to put yourself through this torture. If you lower your allocation when prices go insane, you thereby keep your personal risk profile stable when the risk of stock investing goes through the roof. You lose less in the huge bear market. That leaves you with more to put in stocks when valuations return to reasonable or better levels and the likely long-term return of stocks goes to levels far higher than the average 6 percent number.

    Here’s a study by Wade Pfau,Associate Professor of Economics at the National Graduate Institute of Policy Studies, comparing Valuation-Informed Indexing and Buy-and-Hold:

    Pfau finds that VII provides more wealth for 102 of 110 of the rolling 30-year time-periods in the historical record. That’s while taking on dramatically LESS risk! Higher returns at dramatically lowered risk — This is Investor Heaven!

    And it all conforms to common sense. Price matters with everything else we buy. Why should anyone want to believe that it wouldn’t matter when it comes to buying stocks?

    I can tell you that there has never been a single study showing that valuations do not affect long-term returns. The statement “timing always works” is every bit as true as the statement “timing never works.” It all depends on what form of timing you are talking about. Short-term timing NEVER works. Long-term timing ALWAYS works. This has been so for 140 years now.


  10. says

    Hi, thanks for writing about those 1.8% and 7% numbers. Please let me provide some clarifications about those. I’m copying and pasting this from my blog:

    An article I wrote with Michael Finke and Duncan Williams of Texas Tech University appears in this issue. It is “Spending Flexibility and Safe Withdrawal Rates.” Traditional safe withdrawal rate research focuses on finding the highest withdrawal rate possible with a sufficiently small probability of failure. This ignores the tradeoff that lower withdrawal rates means less spending and less enjoyment in retirement. The more you spend, the more you can enjoy your early retirement, but the higher is your chance of running out of funds later in retirement. We try to balance this tradeoff by considering retirees who have different amounts of external income floors (such as Social Security) and different degrees of personal flexibility about how much spending fluctuations they are willing to endure.

    The findings we describe are starting to get discussed at internet discussion boards and personal finance blog sites. This is primarily because Glenn Ruffenach wrote a column at SmartMoney juxtaposing the findings of this approach with the 1.8% withdrawal rate I described last August in the Journal of Financial Planning. The comments are along the lines of: these moronic researchers can’t figure out if the safe withdrawal rate is 1.8% or 7%. A think a lot of confusion stems from this throwaway line in the introduction of Mr. Ruffenach’s column: “a safe withdrawal rate for some individuals could be as much as 7%.”

    That 7% appears in Figure 3 of our article. But it is not correctly described as a safe withdrawal rate. As we add in this article, this 7% withdrawal rate has a 57% chance of failure over a 30-year retirement. It’s not safe. But what is does is maximize the overall expected lifetime satisfaction for a fairly flexible retired couple who has a secured income base of $20,000 from Social Security. This is how the couple best balances spending more early in retirement with the tradeoff that they may have to spend less later in retirement.

    • MMD says

      Wow! I’m humbled to have one of the people who ultimately contributed to the numbers that appeared in the Smart Money article that prompted this post. Thank you for commenting! Your clarification is indeed helpful. After hearing what you have to say, I’m not sure if I would call 7% safe either.

      Your point about adding other income streams is definitely important but does add a level of complication to the matter. For example, my wife and I plan to have at least $20K in Social Security income. So we may be able to take out more in the early years. But should we? What if we plan to retire early and will need our money to last longer, or what if the Social Security program heads into more trouble and gets cut even further? I realize these are unique questions where the answers are either personal or unpredictable, but they should be considered. Maybe a good suggestion for those seeking defense (as I often look to do) would be to expand upon the importance of various income streams during retirement.

      Perhaps the one-size fits all number doesn’t exist. My personal preference with the conservative 3% suggestion is simply my best guess based on how I feel the market will perform. You can experiment with my model yourself with my Monte Carlo simulation in the following post “A Better Way to See If You’ll Run Out of Money During Retirement”.

      • says


        Thanks. It’s nice to see you are investigating Monte Carlo. About time horizon, if it is different from 30-years, well then the situation is different. I had an article in the January 2012 Journal of Financial Planning in which I prepared a table that gives my best effort at providing some guidelines about a wide variety of situations using Monte Carlo. I summarize the issue here:

        About Social Security, yes if you retire before you begin collecting, then the analysis becomes more complicated and there are no generalized solutions, but you can work something out. You can spend more prior to beginning Social Security with the understanding that you will then cut back a lot after Social Security. And generally, it may benefit retirees to wait until 70 to begin collecting Social Security, but everyone’s situation is different. You can probably work that into the spreadsheet you are developing.

        You might like to plan in mind that Social Security is reduced, but I am not too worried about that for current or near retirees.

        • MMD says

          Wade: Outstanding study! Thank you for sharing this link. These tables do provide some very valuable insight. As you might guess, my eyes were drawn directly to the 40 Year column for both Tables 2.3 and 2.3A. I was pleased to see that in both situations (depending on allocation and acceptable failure) that the withdrawal rate was between 2 and 4%. (By pleased, I meant that my guess of 3% is probably not far off. I am not pleased that I will only be able to take out such a small amount during retirement if I am to have a high probability of success!)

          I am also of the mindset that we will have some form of Social Security in the future, but it was likely be reduced (possibly beyond the 24% reduction my Social Security statements are estimating). Ultimately, I feel like if Social Security were ever entirely dropped or cut altogether, America would have some BIG problems beyond just Social Security.

  11. says

    I started to live off my portfolio in 1999. The research I read that came with the 4% withdrawal was making 8% return and with 3% inflation.

    I started with a 4% withdrawal, but have since gone lower. What I was worried about was that I was taking out some capital as my dividend income was running at 3 to 3.5% per year. I now withdraw 3%. My portfolio has grown at 4% per year and my current dividend income is 3.5%.

    Some think of me as a buy and hold investor, but I look at my investing more as a long term investor. I still have some stocks I bought back in the early 1980’s. But other stocks I have bought and sold. I generally sell when a stock is not performing in the manner I expected when I bought it.

    • MMD says

      This is exactly what I was hoping for: Feedback from someone who is actually retired and can share their experiences with the withdrawal rate. Thanks for commenting! I have mixed feelings about hearing you are at 3%. On one hand I’m happy to hear that my assumption may be correct. On the other hand I’m sad to hear that I may ONLY be able to withdraw 3% and will need a whole lot more money to meet my retirement cash flow target!

      Just so I’m clear, is that 3% from your retirement portfolio and 3.5% from your dividends for a total of 6.5%? Also (off the subject), did you retire before age 59-1/2? I’m intrigued to find out if the lowering from 4% to 3% had anything to do with age?

      • Jim Doherty says

        I am presently learning about retirement withdrawl rates in anticipation of retirement in the next few months. I am 64 with a $1M portfolio mostly invested in dividend paying stocks yeilding approximately 3%. Question? Is the withdrawal rate about which the articles speak against principal excluding income? My thought is to spend the income and have a withdrawal against principal of 2%. Would this be considered to be conservative? Thanks for any insight you can provide Jim

        • MMD says

          Hi Jim: Thank you for stopping by and for asking. I’m no financial advisor, but I’ll try my best.

          The withdrawal rate is not inclusive of your dividend income. For example, if you think you will need $50K per year in retirement, then you would take your 3% in dividend income ($30K) and withdraw 2% of the principal ($20K). Although this will drop your principal down from $1M to $980K, in theory your money should continue to grow (depending on your investment strategy and market performance) around 6% to 8% which would raise it back up to around $1.038M to $1.058M. Therefore, year after year, the process could theoretically continue for the rest of your life. Remember, though, that to say 6 or 8% is just a generic assumption and that there is a danger in assuming your money will always grow year after year. To get more realistic, you may want to also try my Monte Carlo simulation in the next post:

          This simulation will create multiple scenarios where the market is up and down year to year. It’s a clever way to see how your money would do under these types of conditions and if you would ever hit bottom.

          In general, I think 2% is a very conservative and very safe rate to pick. In my opinion, I’d rather shoot low and have my money last than have it the other way around.

          Good job building a $1M portfolio and making it to retirement!

          • Jim Doherty says

            Thank you so much for your input. There is so much written on this topic it is difficult to know how to proceed. It is great that you were willing to share your thoughts! Thanks a Bunch!!

            • MMD says

              My pleasure! Just remember to always be re-evaluating your situation. If things are going great with your dividend income and withdrawals, then good – no change. If the economy changes and your principle sees a significant reduction, then it may be time to revise the strategy.

  12. says

    The lowering of withdrawals from 4% to 3% had to do with the fact I was taking out more than my dividend income and I had had a bear market just after I stopped working. I set about to increase dividend income and only take out dividend income.

    Last year my dividend yield was 3.5% and total withdrawal was 3%. That means that my total withdrawal was less than my dividends. (I track everything on spreadsheets and Quicken.) The thing is my dividend income has been increasing much faster than my budget increases. My yearly dividend increases range from low of 5% to high of 24% with a 5 year median value of 11.4%.

    One of the original financial bloggers I read talked on and on about dividend paying growth stocks. These companies tend to pay dividend around 1%, but they also tend to increase dividends by 15% to 30%. With some of these and other dividend stocks my dividends in 1999 was low, around 2.8%. To increase dividends I bought some companies that paid a higher dividend and raised my dividends income. So now I have a mix dividend stocks with low, medium and high yields and low, medium and high growth in dividends.

    Do not forget with the earn 8% and take out 4% scenario you will be able to take out an increasing amount of money each year. I would not expect to earn 8% returns over the short term and but I do expect to do so over the long term.

    • MMD says

      Thanks for sharing your experiences. Buying dividend stocks this year is on my to-do list. Although the returns seem small at first glimpse, I have read a few books now (as well as some great blogs) about the benefits of going with dividend stocks (like as you mention the increasing payout, the strength of the company, the probability of the stock being a better performer, etc). I was really hoping Apple would surprise us all and payout a dividend in the February meeting, but no such luck!


  1. […] When planning your retirement, you need to think about your withdrawal rate.  My Money Design discussed Is 2% the New Safe Retirement Withdrawal Rate? […]

  2. Planning and Investing Articles from Finance Blogs | Personal Investment Management and Financial Planning says:

    […] Planning MMD presents Is 2 Percent the New Safe Retirement Withdrawal Rate posted at MyMoneyDesign, saying, “Are you planning to withdrawal 4% from your nest egg in […]

  3. […] MMD @ MyMoneyDesign writes Is 2 Percent the New Safe Retirement Withdrawal Rate? – Are you planning to withdrawal 4% from your nest egg in retirement? What if I told you some analysts think 2% may be a more appropriate figure. Let’s calculate the numbers and see for ourselves. […]

  4. […] MMD @ MyMoneyDesign writes Is 2 Percent the New Safe Retirement Withdrawal Rate? – Are you planning to withdrawal 4% from your nest egg in retirement? What if I told you some analysts think 2% may be a more appropriate figure. Let’s calculate the numbers and see for ourselves. March 12th, 2012 | Tags: carnival of retirement, retirement | Category: Uncategorized […]

  5. […] @ MyMoneyDesign writes Is 2 Percent the New Safe Retirement Withdrawal Rate? – Are you planning to withdrawal 4% from your nest egg in retirement? What if I told you some […]

  6. […] My Money Design asked Is 2 Percent the New Safe Retirement Withdrawal Rate? […]

Leave a Reply

Your email address will not be published. Required fields are marked *

CommentLuv badge