Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion

Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.

    Support MFO

  • Donate through PayPal

«1

Comments

  • edited March 2016
    @Shippwrecked, Good article and the following rings true: "It’s actually not the average that matters though. Rather, it’s when the ups and downs happen that determine how likely your portfolio will survive longer."

    We don't take a lot out, but have been traditionally taking the full year's distribution in January. Worked well up until this January. Got stung a little. In the future we'll stagger withdrawals over a full year.
  • Interesting work, partly new and certainly confirmatory. Does anyone know of any scholarly work w backtesting for equity portion (whatever % it is) omitting smallcaps?
  • The rate is important, but living within ones means in retirement is even more important. I have said it many times before, and I say it again: Starting retirement with no mortgage and no burdensome credit card debt can be the most important goal of retirement planning. This alone can reduce spending by 30-50% and makes a huge impact on the withdrawal rate.
  • beebee
    edited March 2016
    I would like to hear from other retirees who have successfully navigate a portfolio's downside risk in retirement from the point of view of a safe withdrawal "method". In other words, along with a SWR (Safe Withdrawal Rate) there also needs to be guidance on managing a SWM (Safe Withdrawal Method). What part of the portfolio will serve as the vehicle for this safe withdrawal "rate" and when will that withdrawal happen. Cash seem to be a part of a retiree portfolio that might help with this.

    @hank lamented as to the "when" with his comment,"We don't take a lot out, but have been traditionally taking the full year's distribution in January. Worked well up until this January. Got stung a little. In the future we'll stagger withdrawals over a full year." As a follow up question, what did you decide to redeem; stock, bonds, or cash?

    Obviously redeeming and withdrawing shares of equities that have lost 50% of their value (due to market volatility) doesn't sound like a very safe withdrawal method, but holding cash (or a cash like investment) as the withdrawal vehicle would seem to help remove market volatility out of a portfolio withdrawal.

    I have found that saving into market volatility (both up or down) isn't nearly as a emotional as spending down a portfolio's assets. Separating out a portion of one's portfolio into cash for emergencies or for future spending might be one way to calm the emotional side of the withdrawal.

    Thanks for the thread @shipwreckedandalone...Additional thoughts?
  • Different issue but in the ballpark in answer to bee's question......I have personally known 4 people who chose a lump sum option from their employer in retirement vs annuity option. All 4 are in serious trouble now. I have known 2 who chose the annuity option and they are spending the entire check Jimmy Buffet style on the beach because they know another check will be in the mailbox in 30 days. This brings up a separate set of questions which I will bring up in a separate discussion thread soon.
  • Bee, I believe the less movable parts the better for a portfolio which is why I like 60/40 to 50/50 funds which balance it for us and make withdrawals easier. I have learned thru the years there are layers and dimensions of risk far beyond the day you buy any other alternative and your post hit on one of those. It seems investors are always confronted with a decision to make of some unpredictable origin if other strategies are adopted and each one of those decisions can be wrong which wipes out all the previous right decisions.
  • @shipwreckedandalone: I couldn't agree more with respect to the "layers and dimensions of risk far beyond the day you buy". Maybe not "always", but for sure a large percentage of the time "investors are... confronted with a decision to make of some unpredictable origin." Murphy is alive and well.
  • Hi Guys,

    Whenever a MFO discussion on retirement planning and drawdown schedule is initiated, my contributions are predictable and fairly consistent. Sorry about that, but I’m a firm believer that Monte Carlo methods are especially appropriate tools to provide actionable guidance.

    I believe I posted on this subject recently, but I’ve forgotten the Discussion title. So I will repost my comments as follows:

    “Simple heuristics (rules-of-thumb) are fine when making common everyday decisions like buying a hamburger or not, but are totally inadequate when making complex, significant decisions like those about retirement.

    The retirement when, where, how much do I need, drawdown rate, portfolio size and placements seem hopelessly intertwined to permit a comfortable and confident decision. But a financial tool is readily accessible that significantly attenuates doubt, and it’s not rule-of-thumb based.

    I’ve proposed this approach many times on MFO, but I don’t hesitate to do so once again. That tool is Monte Carlo simulation analyses. I do not apologize for being a broken record in this instance.

    Many such tools are easily accessible for free on the Internet. Two such codes that I have previously recommended are the PortfolioVisualizer and the MoneyChimp codes. Here are direct Links to these Monte Carlo simulators:

    https://www.portfoliovisualizer.com/monte-carlo-simulation

    http://www.moneychimp.com/articles/volatility/montecarlo.htm

    Please give them a few tries. The inputs are self-explanatory, and the codes are fast. Many scenarios can be explored over a short commitment of time. Endless what-if scenarios can be examined with end portfolio average value and portfolio survival likelihoods as their primary outputs. Thousands of cases are randomly constructed for the projected market returns.

    The PortfolioVisualizer tool has more user options, but the MoneyChimp version also does yeomen work. Since these are Monte Carlo-based codes, each time a simulation is made, expect slightly changed predictions. That somewhat captures the fragile nature of the uncertain future.

    Retirement decisions will be dramatically improved by application of these simulators. Imperfect analyses (even estimating the range of possible market returns is risky business) almost always beats poorly informed guesstimates. Before making a retirement decision, give the Monte Carlo codes a test ride. They are powerful stuff for everyone.

    And for normal circumstances and drawdown rates, a 2 million dollar portfolio is not necessary for a portfolio with some equity holdings. Do the analyses to challenge the robustness of that statement.”

    I hope my repost is helpful to some newer MFO members. Portfolio volatility degrades end wealth. That’s why when constructing a portfolio one goal is to minimize its standard deviation (volatility). Low component standard deviations and low component correlation coefficients work to accomplish that goal.

    A simple equation demonstrates the need to minimize portfolio standard deviation to achieve a higher cumulative return. Cumulative annual return is roughly equal to average annual return minus one-half times the square of the portfolio’s standard deviation. Note the minus sign. Standard deviation always operates to reduce average annual returns over the years.

    Good luck and good planning for your retirement, and for the likelihood of your portfolio’s survival.

    Best Wishes.
  • edited March 2016

    I have personally known 4 people who chose a lump sum option from their employer in retirement vs annuity option. All 4 are in serious trouble now.

    Ya I know someone who did the same thing when he retired 5 years ago at 62 and is in trouble now.
    Spent every dime. Really sad to see someone work 30+ years at a tough job and end up with so little in retirement.
    Maybe our schools need to teach the kids a jingle (to the tune of a Dinsey song I remember as a kid):
    "D-I-S-C-I-P-L ... I-N-E spells Discipline." (dumb - I know)

    Gets back to BobC's comment too about avoiding credit card debt.
  • Speaking of schools and discipline I was on a first-name basis with Al Doyle, the "dean of discipline" in high school. We didn't talk much about economics, though.
  • edited March 2016
    Old_Joe said:

    Speaking of schools ... I was on a first-name basis with Al Doyle, the "dean of discipline" in high school.

    I'll bet you were. How do you spell intractable?
    :)
  • I can do that. I had to write "I will not be an intractable clown" 1499 times!
  • @ Old Joe .. If you were a kid today you would just write a little script telling your computer to print it out 1499 times and watch it spill paper on the floor.......
  • MJG, is not a high number of rolling returns and periods sufficiently MC-like?
  • Hi Davidrmoran,

    Using a set of rolling market returns is a step in the right direction over the original Trinity study approach, but it is a very small step. It still only reflects the exact sequence of returns that were historically registered by the marketplace.

    Market returns will never repeat that precise order. All evidence points to a purely random series of returns. Monte Carlo methods capture that random characteristic. By running 1000 randomly selected return sequences, a user gets a better feeling for the spread in possible outcomes by an order of magnitude or so.

    Given a withdrawal schedule, Monte Carlo projects the likelihood of portfolio survival for elapsed time and projected market return stats. Alternate scenarios and drawdowns are easily explored if the survival prospects are not attractive. I used Monte Carlo as one tool in making my retirement decision.

    Thank you for asking.

    Best Wishes.
  • I will have to compare outcomes, but statistically this may be a distinction without a difference.
  • My experience with hundreds of clients over the years (no matter how many projections we run prior to retirement, MonteCarlo or not) is that those with public pensions (after working 30+ years) seldom have spending problems. The public pension system is very generous, and it allows folks to retire with most of their pre-retirement income continuing. If they also have no mortgage and other heavy debt, they are even in better shape. If the spouse has good social security benefits, even better. With these folks, the withdrawal rate on their other savings is not much of an issue.

    For other clients, our experience has been that folks tend to spend less following down years for the markets, then discover that some of the "necessary" spending they did previously is not so necessary any more. The first 4-5 years of real retirement are when folks do the most traveling and other unusual expenses. But even then, with the exception of those who have always lived beyond their means, in the last 2-3 years folks have been more aware of their spending. As I have noted previously, those with no mortgage and other debts always seem to worry less than those with debts. And for good reason...their expenses without those things are almost always 30-60% lower.
  • beebee
    edited March 2016
    Has anyone explored the tax considerations with respect to Safe Withdrawal Rate?

    I'm gonna assume that a Roth withdrawal could potentially be 15-40% smaller due to the fact that these withdrawal don't get a "income tax haircut" prior to fulfilling their primary need (spending).

    Also, if I am required at 70.5 to take a Required Minimum Distribution, than a $1,000 RMD will only meet a $600 - $850 of my income need (after taxes).

    @MJG: Do you know if RMD is part of a Monte Carlo Simulator? If a retiree is required to take a distribution, pay taxes on that distribution and then either spend or save that distribution would throw a small wrench into MCS.

  • Allow me to question a few assumptions (or not quite precise statements) to refine some thoughts.

    @MJG: "Market returns will never repeat that precise order. All evidence points to a purely random series of returns."

    Never? If results are purely random then there is a nonzero probability of repeating any given pattern (at least over a finite space of possible outcomes).

    I think you're suggesting that there is no persistence, there are no market cycles. In the fixed income market, we have had a 35 year bull market (give or take). All expectations are for an ensuing bear market. Not to mention basic arithmetic - yields rising from virtually zero entail falling security prices.

    How does one square this (both past market and future expectation) with pure randomness? While Y2Y returns may be random, there appears to be a longer term trend that would belie "pure" randomness.

    @BobC - you write about people with public pensions. While private pensions are much rarer these days, they do exist. Is there some difference between public and private sector (all else being equal) or is it that you just don't run across clients with significant private pensions?

    Regarding lump sum vs. annuity (pension) option: I question cause and effect. Do people who take lump sums fare worse because they take lump sums, or is it that people who do not manage money well are more inclined to take lump sums and spend?

    @bee IMHO RMDs, if properly managed, should not be a factor. People tend to look at pretax portfolios and think that they have $X. But what they really have is $X * (1 - tax). Taking RMDs doesn't change this.

    I say "properly managed" because letting pretax accounts pile up can ultimately push one into higher tax brackets. This reduces the after tax value more than necessary. By spreading the withdrawals over time one can reduce the tax impact and maximize the aftertax value of the account.

    If one doesn't need the full amount of the withdrawals, one can put the excess into Roth IRAs and prolong the tax sheltering.
  • Hi msf,

    You are absolutely correct mathematically that "Never" is too strong when events are random. The odds of an exact replication of past returns is not precisely zero, but the odds approach zero, especially as the timeframe expands. I should have qualified my "Never" statement with something like "almost never" or "approaching never".

    Markets must promise a return in excess of the expected inflation rate. Otherwise, investments would not be made. The market's must have an upward bias to attract investors. Randomness can exist and does exist with a positive non-zero average return.

    Best Wishes.
  • Newer academic thinking about investment glide path allocations and withdrawal rates in retirement years ( Weigand and Iron / Sptizer and Singh *) has shown that an investor / retiree spend from bonds first and stocks last ( and build a "safe money" fund or bucket of approx. 2 years of expenses which can be used if needed or spent before bonds ). Under this thinking, a misconception about conventional 60 / 40 "glide path" schemes is, that a "bond" allocation be recommended "early" in the investment lifecycle. Yet, the young investor demographic ( age 20's to 50 ) has "time" compounding / "time" to ride out volatility advantages on their side and they aren't so invested in knowing the quarter to quarter fluctuations of their 401K portfolios. So it is logical to assume that a "maximizing" of asset growth by having a much higher portion of assets in equities is warranted and, consequently, should extend into an investors "final years".


    Being a late 50's retiree with a somewhat limited but reasonable Roth IRA accumulation and with an extensive expertise in quantitative tactical allocation, I operate under the framework of "preservation of capital" model with an appreciation of what the Weigand and Iron study conveys. As the forward 15 year equity market returns, as measured by CAPE ** and price to book measures are extrapolated to be sub par, preserving capital and asset growth within alternating strategic periods of equity ( small cap value, mid cap growth ), money market, and occasional bond investment through the use of quantitative tactical methods, is my preferred choice. Many "equities heavy" buy and hold investors / retirees may have to ride out the overvaluation period, perhaps spending down their safe money portion and/or retirement asset stake, as is implied by "sequence of return risk". The unknown is how deep and how long the overvaluation period is; this accompanied by varying inflation / disinflation .
    Historically, a simple, mechanical, low transaction price / moving average cross strategy has produced decent risk mitigation / capital preservation during these periods of CAPE overvaluation ***.

    Some favorite quotes from retirement planner literature are: "Hope for the best, plan for the worst", "You can't predict, but you can prepare ".



    * "Market Signals for When to Employ a Bonds-First Withdrawal Sequence to Extend the Longevity of Retirees’ Portfolios" R. Weigand
    "Is Rebalancing a Portfolio During Retirement Necessary?" John Spitzer Sandeep Singh
    ** https://docs.google.com/document/d/1I4sH5UV6fS6UfCNiPl1AsB2SOMF1an1PRt8YH0dgOeQ/edit?usp=sharing
    *** https://docs.google.com/presentation/d/1mdon_cto48rvs2_lKWyMWrfqSIh8K0phfe7tThle8qQ/edit?usp=sharing
    https://docs.google.com/presentation/d/1Sn6BKRCKRU5tensBDFTkJXI3v2wRQ4M1bt8VoIM2Zmc/edit?usp=sharing
  • In a stripped down form, I believe this is what one gets out of Buffett's advice for his future widow - 10% short term government bonds (effectively a cash substitute) and 90% in an equity index fund.
    https://blogs.cfainstitute.org/investor/2014/03/04/warren-buffetts-90-10-rule-of-thumb-for-retirement-investing/

    That's 2.5-3 years of buffer. As you noted, the length of the overvaluation period (do you mean undervaluation when a retiree is disinvesting?) is unknown, but that is likely enough to insulate one from the worst of it. If that period extends further, one does not need to replenish the buffer, merely sell off enough equity to meet cash flow needs. Once stocks return to a reasonable valuation level, the buffer can be refilled.

    Personally, I'm more comfortable with a 4-5 year buffer and a more diversified equity portfolio, but generally find this a good approach.
  • beebee
    edited March 2016
    Interesting Read using a three fund portfolio (VFINX, VUSTX, VSGBX or VFITX) and a 200 mda filter.

    From the link:
    "The popular 60/40 Stocks/Bond portfolio performs well over the past 24 years, but adding a simple moving average to this portfolio has increased returns, reduced the duration of draw downs, and substantially reduced portfolio draw down. Adding in an intermediate term bond fund as the cash fund accomplished even more, it increased annual returns more than 10% over the buy and hold portfolio, while having close to 1/3 of the daily draw down numbers. Avoiding draw down and still being involved in market upswings was the goal of this strategy, and it worked well in this instance. There were a few concerns, namely being involved in the cash filter fund for too much duration, not being diverse enough to capitalize on gains across different markets, and the potential of missing out on some of the market upsides. However, these concerns did not prevent us from accomplishing the goals of reducing draw down and risk along with increasing return in this particular example."

    iema-blog.com/2016/02/6040-stockbonds-portfolio-with-market.html
  • Hi Bee,

    You asked: "Do you know if RMD is part of a Monte Carlo Simulator?"

    I can give an unconfirmed "yes" to your question. It is available on my earlier referenced PortfolioVisualizer Monte Carlo code. I say unconfirmed because I've never used that option.

    That Monte Carlo simulator has an input option titled Life Expectancy Based Annual Withdrawal". It purportedly models a RMD drawdown requirement.

    That option is only operational starting at age 70. If you want to run a simulation starting at an earlier age, you'll need to run it in two parts. Run the code from your current age of interest until 70, and next run it again starting at age 70 with inputs from the earlier simulation's output.

    I hope this helps and I hope you exercise the flexibility that the PortfolioVisualizer tool offers. Sequentially running that tool expands its flexibility.

    Best Wishes.
  • beebee
    edited March 2016
    Getting back to the question as to "where" should retirement withdrawals come from this study researched a number of options and I liked this quote enough to pass it along:

    In virtually all the scenarios, "it pays to eat your bonds first, equities later."

    Withdrawal scenarios studied:

    1. Withdraw money from either stocks or bonds and then rebalance the portfolio annually to the initial stock/bond proportion. This harvesting rule will be referred to as “Rebalance.”

    2. Withdraw money from the asset that had the highest return during the year and do not rebalance. This will be referred to as “High First.”

    3. Withdraw money from the asset that had the lowest return during the year and do not rebalance. This will be referred to as “Low First.” To the extent that historical rates of return on bonds tend to be lower than historical rates of return on stocks, the following two additional methods of harvesting withdrawals will be referred to as “Bonds First” and “Stocks First.”

    4. Take withdrawals from bonds first and do not rebalance.

    5. Take withdrawals from stocks first and do not rebalance.

    Study:
    time-diversification-vs-rebalancing-in-retirement-portfolios/
  • bee said:

    Getting back to the question as to "where" should retirement withdrawals come from this study researched a number of options and I liked this quote enough to pass it along:

    In virtually all the scenarios, "it pays to eat your bonds first, equities later."

    Withdrawal scenarios studied:

    1. Withdraw money from either stocks or bonds and then rebalance the portfolio annually to the initial stock/bond proportion. This harvesting rule will be referred to as “Rebalance.”

    2. Withdraw money from the asset that had the highest return during the year and do not rebalance. This will be referred to as “High First.”

    3. Withdraw money from the asset that had the lowest return during the year and do not rebalance. This will be referred to as “Low First.” To the extent that historical rates of return on bonds tend to be lower than historical rates of return on stocks, the following two additional methods of harvesting withdrawals will be referred to as “Bonds First” and “Stocks First.”

    4. Take withdrawals from bonds first and do not rebalance.

    5. Take withdrawals from stocks first and do not rebalance.

    Study:
    time-diversification-vs-rebalancing-in-retirement-portfolios/

    Yes, that was the Spitzer and Singh study cited ...
    bee said:

    Interesting Read using a three fund portfolio (VFINX, VUSTX, VSGBX or VFITX) and a 200 mda filter.

    From the link:
    "The popular 60/40 Stocks/Bond portfolio performs well over the past 24 years, but adding a simple moving average to this portfolio has increased returns, reduced the duration of draw downs, and substantially reduced portfolio draw down. Adding in an intermediate term bond fund as the cash fund accomplished even more, it increased annual returns more than 10% over the buy and hold portfolio, while having close to 1/3 of the daily draw down numbers. Avoiding draw down and still being involved in market upswings was the goal of this strategy, and it worked well in this instance. There were a few concerns, namely being involved in the cash filter fund for too much duration, not being diverse enough to capitalize on gains across different markets, and the potential of missing out on some of the market upsides. However, these concerns did not prevent us from accomplishing the goals of reducing draw down and risk along with increasing return in this particular example."

    iema-blog.com/2016/02/6040-stockbonds-portfolio-with-market.html

    They could probably run a 10 period monthly moving average on the prices of the assets as to reduce daily generated "whipsaws" ( as many "needless" whipsaws occurring in the past have been contained "within" the monthly data ) and reduce the amount of "management" time, ie. looking at the calculations daily / subjecting oneself too frequently to market data - leading to possible cognitive investing biases ...

    Also, using healthcare for the 60% allocation has produced alpha ( appreciably ) above VFINX ( 13% CAGR, Sharpe 1.0 -21% max DD with non MA strategy / rebalance annually 1986 - 2015 )
  • Many retirees choose to take the lump sum option with the thought in mind to pay off all debt because they say it is impossible to do with the annuity option. Therefore reduced principal right out of the gate. Many are not sophisticated investors, many retired before the great recession, many have withdrawal rates too high...approved by financial advisers who know they will not get the assets unless they approve the high withdrawal rate. So they are way down the wrong path within two weeks of retirement.
  • beebee
    edited March 2016
    Debt is a drag on one's available resources.

    If a newly minted retiree has had the discipline to pay off debt prior to retirement more power to them, but using your retirement nest egg to eliminate debt at the start of retirement seems counter intuitive to why these dollars where saved in the first place. If eliminating debt is a priority, keep working, stop saving for retirement and eliminate debt with working income, not your retirement nest egg.

    A very wise retiree once told me that prior to retirement try to live on your retirement income...save the extra or use it to pay down debt. One quickly realizes if they are ready to consider living on retirement income. If not, no harm...no foul...keep working.

    Also, I'm a big advocate of maintaining a monthly financial spreadsheet. Track everything and you will soon develop a picture that can be very helpful in decision making.
  • edited March 2016
    I have been a bit obsessed recently with retirement - which means less trading. I have given deferred annuities a close look including the newer QLAC's. As much as I try though, I just can't get into them. For the most part you just get your principal paid back to you until the age at which the longevity tables say you are suppose to expire. They only seem to be of benefit to the insurance companies and those who live to a ripe old age.

    When and if rates ever rise, I can see me putting as much as 25% of my nest egg in 5 year CDs to offset part of my living expenses (after allowing of course for any income from SS, CDs, etc.) My biggest expense in retirement will be the taxes paid on my RMD.

    As for withdrawal rates. I look at my nest egg and then my estimated annual spending in retirement (and err on the very high side) It's not like I have some huge oversized nest egg. But I don't need Monte Carlo analysis to tell me that even if I were 100% in cash forever, that I better start spending more money now and enjoying life lest I die with too much of a nest egg.

    In retirement it helps to be single, live in a very low cost living area of the country, be a frugalist, and be 100% debt free. I have those bases covered. Then again, it isn't fun to be alone in old age. But I also have that covered as my long time lady friend is my neighbor.
Sign In or Register to comment.