Trying to Better Understand Safe Withdrawal Rates

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.

Trying to Better Understand Safe Withdrawal Rates

Postby Valdeselad » Mon Jul 15, 2013 4:06 pm

All,

I'm trying to get a better understanding on the details and assumptions of longer-term safe withdrawal rates. When a comment is made referring to 2, 3 or 4% SWR, is this supposed to be valid for an infinite amount of time or only for a set # of years (wiki has 15-30 years)? Does this base SWR assume a certain AA (i.e. 60/40)? For example, how do you know if the SWR applies to someone at 50, 60 or perhaps 70 years old? What about someone younger than this? I have read the wiki and, while helpful, this is one of those topics which I can't quite become comfortable.

My hesitation keeps coming back to the fact of not knowing when enough is enough. Let's say you retired at 55 with a nest egg of $4 million dollars and assumed a 4% SWR which would give you over $150k/year of income. Should a person assume that (if they make it) when they turn 95, their portfolio would be running on empty or still flush with cash? If it's still full of cash, perhaps you could have retired 10 years earlier. If it's empty, perhaps you should have waited a few more years before hanging it up. In the end, it just seems like an educated guess based on your long-term expense profile - one which you probably should be ultra-conservative with your assumptions. What am I missing? How do the safe withdrawal guidelines really help?

As for me, I take something of a hybrid approach. I have a long-term income statement showing my current (and estimated future) expenses and retirement balances. I also have a line showing the estimated (conservative) yield of my portfolio. There is a point in the future where the estimated yield of my portfolio basically covers what I think would be my core annual expenses. This is also right around the time where a 4% withdrawal rate from my portfolio would cover about 3x my estimated annual expenses, providing what I hope would be a ultra conservative cushion in case the market returns for the next 40 or 50 years don't match the last 40 or 50 - Hope for the best but prepare for the worst I suppose.
Valdeselad
 
Posts: 112
Joined: Mon Apr 26, 2010 2:42 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby gvsucavie03 » Mon Jul 15, 2013 4:34 pm

Wiki article link: Safe Withdrawal Rates scroll down to Papers and read up on the "Retirement Calculator from Hell." I read through this and many of the other studies on SWR and some similarities surface:

A 4% fixed withdraw rate is very likely to succeed over a 30-year period with around a 50/50 portfolio or more tilt towards stocks (although much more volitle as you add more stocks, but you probably won't run out of money as if you had a 100% FI portfolio). The Retirement Calculator from Hell was based on the 1965-1995 market where there was a long bear market at the start of retirement and no significant gains were made until much later. A 4% variable withdraw rate (income based on the portfolio balance each year) probably has more of a chance for success, but your annual investment income might cut by half or more. Stock ownership in all research I've seen is a must for sustaining a portfolio. The premise of this study was that online retirement calculators don't factor in the fluctuation of returns on a portfolio in retirement (especially with the chance that you may retire into a long bear market like '65 that spent 5-10 years with a real return rate of 0%).

This study, overlaid with other studies on portfolio success with varying SWR's and AA's makes me feel that a fixed 4% SWR with at least a 50/50 AA has a pretty good chance for success.
gvsucavie03
 
Posts: 606
Joined: Sat Feb 16, 2013 9:30 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Valdeselad » Mon Jul 15, 2013 5:10 pm

gvsucavie03 wrote:This study, overlaid with other studies on portfolio success with varying SWR's and AA's makes me feel that a fixed 4% SWR with at least a 50/50 AA has a pretty good chance for success.


But that's for 30 years, right? What would the SWR and AA be if you wanted to be confident for 40 years? 50 years? I think the 4% for 30 years, etc. I'm fine with b/c that assumes a "normal" retirement timeframe. I guess I'm somewhat trying to extend the equation for those looking at a 40, 50, or even possibly 60 year retirement.
Valdeselad
 
Posts: 112
Joined: Mon Apr 26, 2010 2:42 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby gvsucavie03 » Mon Jul 15, 2013 5:33 pm

Valdeselad wrote:
gvsucavie03 wrote:This study, overlaid with other studies on portfolio success with varying SWR's and AA's makes me feel that a fixed 4% SWR with at least a 50/50 AA has a pretty good chance for success.


But that's for 30 years, right? What would the SWR and AA be if you wanted to be confident for 40 years? 50 years? I think the 4% for 30 years, etc. I'm fine with b/c that assumes a "normal" retirement timeframe. I guess I'm somewhat trying to extend the equation for those looking at a 40, 50, or even possibly 60 year retirement.


You'd probably be better with a more aggressive AA (60/40 for example) or lower SWR (3.5 or even 3%) if the 40-50 year retirement is a possibility.
gvsucavie03
 
Posts: 606
Joined: Sat Feb 16, 2013 9:30 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Oliver » Mon Jul 15, 2013 6:48 pm

Valdeselad wrote:
gvsucavie03 wrote:This study, overlaid with other studies on portfolio success with varying SWR's and AA's makes me feel that a fixed 4% SWR with at least a 50/50 AA has a pretty good chance for success.


But that's for 30 years, right? What would the SWR and AA be if you wanted to be confident for 40 years? 50 years? I think the 4% for 30 years, etc. I'm fine with b/c that assumes a "normal" retirement timeframe. I guess I'm somewhat trying to extend the equation for those looking at a 40, 50, or even possibly 60 year retirement.


Before you use 4% as your SWR, you should look at the flaws in the original Trinity study and other research. Personally, I think a SWR below 3% is reasonable. I am using a SWR of 2.75%.

The most recent paper on SWR I know is the following:

Blanchett, David, Finke, Michael S. and Pfau, Wade, Asset Valuations and Safe Portfolio Withdrawal Rates (June 27, 2013). Available at SSRN: http://ssrn.com/abstract=2286146 or http://dx.doi.org/10.2139/ssrn.2286146

PS If the interest rate on TIPS increase, I will be building a TIPS bond ladder. You might google Zvi Bodi. By partially annuitizing my portfolio via TIPS, I can take more money out.
Oliver
 
Posts: 71
Joined: Tue Sep 04, 2007 12:49 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby gvsucavie03 » Mon Jul 15, 2013 7:28 pm

If a portfolio were to average 8-9% over its lifetime and inflation steals 3%, a 4% SWR is still conservative, and plus it has held up to hypothetical worst-case portfolios such as the '65-'95 stretch (as long as a 50/50 AA is maintained).

I think probably the best idea would be just to make the 4% (or 3 or 2.75) a fixed percentage based on the current balance of the portfolio. This is assuming you can tough out a huge dip in your income - as this can certainly happen if the market takes a dive. The 5% SWR held up well in this situation in all scenarios of the "Hell" study that held at least a 50% stock position. A 3% fixed percentage (or less) would likely have seen a large increase and quite a generous inheritance :moneybag

Third option - just withdraw less money if we have another huge crash like in '08, tighten the belt, then let it come back up when the portfolio recovers. Just don't change the AA (make sure it has at least 50% stocks so it can have a better rebound in a bull market).
gvsucavie03
 
Posts: 606
Joined: Sat Feb 16, 2013 9:30 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby ourbrooks » Mon Jul 15, 2013 9:20 pm

gvsucavie03 wrote:If a portfolio were to average 8-9% over its lifetime and inflation steals 3%, a 4% SWR is still conservative, and plus it has held up to hypothetical worst-case portfolios such as the '65-'95 stretch (as long as a 50/50 AA is maintained).

I think probably the best idea would be just to make the 4% (or 3 or 2.75) a fixed percentage based on the current balance of the portfolio. This is assuming you can tough out a huge dip in your income - as this can certainly happen if the market takes a dive. The 5% SWR held up well in this situation in all scenarios of the "Hell" study that held at least a 50% stock position. A 3% fixed percentage (or less) would likely have seen a large increase and quite a generous inheritance :moneybag

Third option - just withdraw less money if we have another huge crash like in '08, tighten the belt, then let it come back up when the portfolio recovers. Just don't change the AA (make sure it has at least 50% stocks so it can have a better rebound in a bull market).


Alas, there's something called the "sequence of returns" effects. Even if stocks return a real 8-9% on average, they also can have negative returns four or five years in a row. If those bad years happen early in retirement, and a retiree sells stocks for income then, the amount that is available for future returns is lower. The numbers you get by adding and subtracting average returns don't take into account these effects. In fact, with an 80% stocks portfolio you're less likely to run out of money than with 100% stocks, even though 100% stocks, on average, returns more!

As to the fixed percentage method, it has one fatal flaw. You never run completely out of money but your portfolio can get arbitrarily small; 4% of $9.95 is probably not enough to live on.
ourbrooks
 
Posts: 1135
Joined: Fri Nov 13, 2009 5:56 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Rodc » Tue Jul 16, 2013 10:39 am

I suspect you desire what you cannot have.

The original Trinity Study has been bent totally out of shape by many (thus the aforementioned "flaws").

At the time there was a prevailing thought that you could pull something like 8% out per year indexed to inflation. The authors of the Trinity Study were just trying to show that was not reasonable. They were not trying to say you should plan on taking out 4% plus inflation come hell or high water. The so-called "4% rule" was never supposed to be a real plan; it was supposed to be a ball park planning figure that would need to be adjusted up or down over time depending on how the portfolio was doing.

And despite the efforts of many to refine that 4% ball park figure, no way can anyone, no matter how fancy their models, give you more significant digits. The honest to goodness answer is that, yeah, if we get really bad times, and someone gets unfortunate timing, it might well be rather less. Likewise, if we get good times, you might be able to take out rather more. Just consider, we have about enough data for 3 quasi-independent 30 year periods. How can you say from three data points what the worst outcome might be? No one knows.

But given a set of volatile, risky assets, there is no guaranteed safe withdrawal rate. And the longer you need the money to last, the greater the chance of terrible showing up along the way.

The closest you will get to a guarantee over X years is to come as close as you can to an X year TIPS ladder. That too of course has its problems.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Rodc
 
Posts: 10114
Joined: Tue Jun 26, 2007 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby HomerJ » Tue Jul 16, 2013 11:25 am

4% is just a rule of thumb for a 30-year retirement with a 50/50 portfolio...

They found that in the past, 4% worked even in the worst of 30-year periods... Even periods of time that include the Great Depression, or periods of time that include a 16 year bear market where stocks went nowhere (DOW was around 1000 in 1966, and STILL at 1000 in 1982 - note that bonds were doing terrible during this time as well because interest rates and inflation were rising)

And STILL 4% worked. Does that mean it will work for all 30-year periods going forward? No... but 4% isn't a "normal times" number... It worked during the really bad 30-year periods in the past.

It's quite likely that you won't retire into the worst 30-year period and you'll have a ton of money left over if you followed the 4% rule... In many 30-year periods, you could easily take 5% or 6% and not run out of money...

But to be safe, it's best to start with 4% and see how things go... To protect yourself from the really bad 30-year periods.

Basically, it's just a rule of thumb, and you adjust your spending each year based on actual events. I won't retire the day 4% will cover all my basic needs... I'd want a little bit of a buffer, so a bad couple of years only means I skip a vacation or two, not that I have to choose between heat or food.

Some people have taken this to an extreme... and claim if you retire TODAY, you better only take 2.8%... which is basically saying that they expect all investments to do worse than inflation for THIRTY years..

If stocks and bonds can match inflation (Currently at 2%), you should be able to pull 3.33% at the minimum for 30 years... I'm not sure why people think 2.8% is a good number... They really think stocks and bonds are going to return less than 2% for 30 years?
User avatar
HomerJ
 
Posts: 6662
Joined: Fri Jun 06, 2008 1:50 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby freebeer » Tue Jul 16, 2013 11:50 am

HomerJ wrote:... Some people have taken this to an extreme... and claim if you retire TODAY, you better only take 2.8%... which is basically saying that they expect all investments to do worse than inflation for THIRTY years..


Good observation!

Also the 30 years is notional, very few 65-year-old male retirees will live to 95, so a longevity-adjusted SWR calculation (/ partial annuitization) would lead to an a higher number.
freebeer
 
Posts: 1278
Joined: Wed May 02, 2007 9:30 am
Location: Seattle area USA

Re: Trying to Better Understand Safe Withdrawal Rates

Postby abhikush » Tue Jul 16, 2013 11:54 am

Oliver wrote:Before you use 4% as your SWR, you should look at the flaws in the original Trinity study and other research. Personally, I think a SWR below 3% is reasonable. I am using a SWR of 2.75%.

The most recent paper on SWR I know is the following:

Blanchett, David, Finke, Michael S. and Pfau, Wade, Asset Valuations and Safe Portfolio Withdrawal Rates (June 27, 2013). Available at SSRN: http://ssrn.com/abstract=2286146 or http://dx.doi.org/10.2139/ssrn.2286146

PS If the interest rate on TIPS increase, I will be building a TIPS bond ladder. You might google Zvi Bodi. By partially annuitizing my portfolio via TIPS, I can take more money out.


Can someone explain table 3 of the paper? It looks like for all probabilities greater than 80% the lower the equity allocation the higher the SWR is, which is counter-intuitive.

For Example, the paper says that in the current environment (Bond Yield 2.0%, CAPE 22), SWR for 95% success is 2.8%, 2.5%, 2.0% and 1.5% for equity allocation of 20%, 40%, 60% & 80% respectively i.e. as you increase the equity allocation the SWR decreases.

Why does SWR decrease when Equity Allocation increases?
abhikush
 
Posts: 1
Joined: Mon Jan 07, 2013 7:04 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Rodc » Tue Jul 16, 2013 12:15 pm

freebeer wrote:
HomerJ wrote:... Some people have taken this to an extreme... and claim if you retire TODAY, you better only take 2.8%... which is basically saying that they expect all investments to do worse than inflation for THIRTY years..


Good observation!

Also the 30 years is notional, very few 65-year-old male retirees will live to 95, so a longevity-adjusted SWR calculation (/ partial annuitization) would lead to an a higher number.


True that it is notional as most of us will not live to 95. However the odds go up a lot if you ask will one of a couple live to 95. Of course it really stinks to be one of the lucky long-lived folks if you happen to go flat broke years before you die. You are not an average.

Also to other points in other posts:

But also, 3.33% (real) over 30 years requires better returns than inflation because of volatility drag, and especially if you get hit with a down period at the beginning. You can't just divide by 30; it does not work that way.

One last thing, the Trinity Study did not show that 4% was always safe in the past, just that 4% gave a very high success rate. A very high success rate means there were failures. And there is only data for about 3 non-over lapping periods and the original paper only had 2. If something works for years 1-30, it is no surprise, nor really additional information, if it works for period 2-31 as there are 28 years in common.

If might be useful to look at what the wiki has to say: http://www.bogleheads.org/wiki/Trinity_study_update
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Rodc
 
Posts: 10114
Joined: Tue Jun 26, 2007 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Oliver » Wed Jul 17, 2013 11:21 pm

One of the papers mentioned in the footnotes of the wiki looks at the results if you replaced the USA data with other countries data. Bottom line - the 4 percent rule does not work in most countries.

Wade Pfau |Pfau, Wade D. 2010. “An International Perspective on Safe Withdrawal Rates from Retirement Savings: The Demise of the 4 Percent Rule?" GRIPS Discussion Papers 10-12, National Graduate Institute for Policy Studies, revised Oct 2010.

http://ideas.repec.org/p/ngi/dpaper/10-12.html
Oliver
 
Posts: 71
Joined: Tue Sep 04, 2007 12:49 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby umfundi » Thu Jul 18, 2013 12:25 am

You can easily guarantee a 4% payment, indexed for inflation, for the rest of your life, no matter how long: Buy a single premium immediate annuity (SPIA).

Other than that, there is really no "safe" withdrawal rate. Any withdrawal has some probability of failure, that you will fail to meet your income needs or even run out of money within 30 or 40 years.

I think you should also consider failure on the upside: You will live more frugally than you need to, and die with plenty of resources left. Even at 4%, this is a much more likely scenario than running out of money.

Keith
Déjà Vu is not a prediction
umfundi
 
Posts: 3361
Joined: Tue Jun 07, 2011 6:26 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby HomerJ » Thu Jul 18, 2013 1:45 pm

umfundi wrote:You can easily guarantee a 4% payment, indexed for inflation, for the rest of your life, no matter how long: Buy a single premium immediate annuity (SPIA).


Something to note... You can always use the existance of SPIAs as a backup plan (if you retire in your 60s - doesn't work for early retirement)...

Start with 4%... Decent chance that your money will actually grow in real terms with that plan.

Now, if you do end up retiring at the one of the worst times in history, you just buy a SPIA at 75 or 80 (10 or 15 years into your "30-year" plan) with the remainder of your money. At that age, you'll get a very good return.

So say you do 4%, and after 12 years, you've lost 50% of your money in real terms... Doesn't look good, right?... but I bet at 77 you can take what's left and get a 8% SPIA which gets you the same amount you were taking out before.

I think you should also consider failure on the upside: You will live more frugally than you need to, and die with plenty of resources left. Even at 4%, this is a much more likely scenario than running out of money.


This is a real risk... Not only just living more frugally than you have to, but working 5-10 YEARS longer than you have to. With a real not insignificant risk of dying 1-10 years into your retirement, working an extra 5-10 years so you can manage a 2.8% withdrawal rate seems more risky than taking a 4% withdrawal rate and adjusting to events as needed.
User avatar
HomerJ
 
Posts: 6662
Joined: Fri Jun 06, 2008 1:50 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Rodc » Thu Jul 18, 2013 1:57 pm

HomerJ wrote:
umfundi wrote:You can easily guarantee a 4% payment, indexed for inflation, for the rest of your life, no matter how long: Buy a single premium immediate annuity (SPIA).


Something to note... You can always use the existance of SPIAs as a backup plan (if you retire in your 60s - doesn't work for early retirement)...

Start with 4%... Decent chance that your money will actually grow in real terms with that plan.

Now, if you do end up retiring at the one of the worst times in history, you just buy a SPIA at 75 or 80 (10 or 15 years into your "30-year" plan) with the remainder of your money. At that age, you'll get a very good return.

So say you do 4%, and after 12 years, you've lost 50% of your money in real terms... Doesn't look good, right?... but I bet at 77 you can take what's left and get a 8% SPIA which gets you the same amount you were taking out before.

I think you should also consider failure on the upside: You will live more frugally than you need to, and die with plenty of resources left. Even at 4%, this is a much more likely scenario than running out of money.


This is a real risk... Not only just living more frugally than you have to, but working 5-10 YEARS longer than you have to. With a real not insignificant risk of dying 1-10 years into your retirement, working an extra 5-10 years so you can manage a 2.8% withdrawal rate seems more risky than taking a 4% withdrawal rate and adjusting to events as needed.


Some also have the option of working part-time or moving a low/pay stress job they really like. Sort of a middle ground. A rather modest income on top of your other retirement income can help a great deal.

Not everyone has that option of course, depends on health, skill set and how the economy is dong.

My dad taught part time until he was 79. He really enjoyed it and the modest income actually made a big difference.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Rodc
 
Posts: 10114
Joined: Tue Jun 26, 2007 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby ourbrooks » Thu Jul 18, 2013 3:28 pm

abhikush wrote:
Oliver wrote:Before you use 4% as your SWR, you should look at the flaws in the original Trinity study and other research. Personally, I think a SWR below 3% is reasonable. I am using a SWR of 2.75%.

The most recent paper on SWR I know is the following:

Blanchett, David, Finke, Michael S. and Pfau, Wade, Asset Valuations and Safe Portfolio Withdrawal Rates (June 27, 2013). Available at SSRN: http://ssrn.com/abstract=2286146 or http://dx.doi.org/10.2139/ssrn.2286146

PS If the interest rate on TIPS increase, I will be building a TIPS bond ladder. You might google Zvi Bodi. By partially annuitizing my portfolio via TIPS, I can take more money out.


Can someone explain table 3 of the paper? It looks like for all probabilities greater than 80% the lower the equity allocation the higher the SWR is, which is counter-intuitive.

For Example, the paper says that in the current environment (Bond Yield 2.0%, CAPE 22), SWR for 95% success is 2.8%, 2.5%, 2.0% and 1.5% for equity allocation of 20%, 40%, 60% & 80% respectively i.e. as you increase the equity allocation the SWR decreases.

Why does SWR decrease when Equity Allocation increases?


Easy answer: With a CAPE of 22, expected returns on equities are very low - around 3% I think - plus equities are more volatile so there's more chance of having a bad run of returns at the beginning of retirement. To compensate for the greater volatility and lower returns of stocks, the SWR needs to be decreased.

Unfortunately, that's a pretty useless result. There's no way of knowing what CAPE will be in the future. A modest market correction of 20% the year after next would lower CAPE to 17, close to its historical median. At the level the expected return on stocks jumps to something like 8% real and increasing your stocks are likely to improve your withdrawal rate. Also, even if prices don't drop, earning can increase. That would also change CAPE and change the withdrawal rate estimates.
ourbrooks
 
Posts: 1135
Joined: Fri Nov 13, 2009 5:56 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Peter Foley » Thu Jul 18, 2013 4:02 pm

From the Bogleheads Wiki:

The income to be derived from an investment portfolio will reflect any monetary amounts you need to regularly withdraw over your life expectancy; any additional monetary amounts you may desire to withdraw if possible; and the taxes you will pay on the withdrawn income. The costs of administering the portfolio are also a critical factor. [2] These withdrawals will provide you with the percentage of your wealth these needs and desires imply and can indicate the degree of flexibility you may have to allow your withdrawals to fluctuate over time. The Trinity and other studies examine the question of what percentage portfolio withdrawal rate (over a number of portfolio asset allocations) is sustainable without being prematurely depleted (the early studies' suggested a sustainable withdrawal rate of 4%).
(my emphasis in bold)

First, regarding the Trinity study itself, the fact that 75% stocks increased the probability of success for 4% inflation-adjusted withdrawals is not a strong reason to choose this allocation. The difference between 100% success for 75/25 and 96% for 50/50 is just that in two years (1965 and 1966) the maximum withdrawal rate fell slightly below 4% with 50/50, but stayed slightly above 4% with 75/25. It is not that big of deal when considering your overall return requirements and tolerance for risk.



Rodc wrote:

The original Trinity Study has been bent totally out of shape by many (thus the aforementioned "flaws").

At the time there was a prevailing thought that you could pull something like 8% out per year indexed to inflation. The authors of the Trinity Study were just trying to show that was not reasonable. They were not trying to say you should plan on taking out 4% plus inflation come hell or high water. The so-called "4% rule" was never supposed to be a real plan; it was supposed to be a ball park planning figure that would need to be adjusted up or down over time depending on how the portfolio was doing.


I agree with Rod. In response to the OP's basic question - the withdrawal rate should take into account one's life expectancy (about 30 years if retiring at age 60) and is based on an asset allocation in a rough range between 40/60 and 75/25.

Changing the components of AA by adding asset classes not available over multiple 30 year periods (TIPs come to mind) would likely impact the 4% rule of thumb for 30 year calculations as well as impact the probability of depletion of assets over longer or shorter periods of time.
User avatar
Peter Foley
 
Posts: 2228
Joined: Fri Nov 23, 2007 11:34 am
Location: Lake Wobegon

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Cut-Throat » Thu Jul 18, 2013 4:28 pm

umfundi wrote:You can easily guarantee a 4% payment, indexed for inflation, for the rest of your life, no matter how long: Buy a single premium immediate annuity (SPIA).

Other than that, there is really no "safe" withdrawal rate. Any withdrawal has some probability of failure, that you will fail to meet your income needs or even run out of money within 30 or 40 years.

I think you should also consider failure on the upside: You will live more frugally than you need to, and die with plenty of resources left. Even at 4%, this is a much more likely scenario than running out of money.

Keith


Keith,

I assume that this is for 1 person, correct? How about a couple Male - 62 years old, Female 54 years old ? ........What would the WR be?
User avatar
Cut-Throat
 
Posts: 1975
Joined: Sun Oct 17, 2010 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby umfundi » Thu Jul 18, 2013 5:13 pm

Cut-Throat wrote:
umfundi wrote:You can easily guarantee a 4% payment, indexed for inflation, for the rest of your life, no matter how long: Buy a single premium immediate annuity (SPIA).

Other than that, there is really no "safe" withdrawal rate. Any withdrawal has some probability of failure, that you will fail to meet your income needs or even run out of money within 30 or 40 years.

I think you should also consider failure on the upside: You will live more frugally than you need to, and die with plenty of resources left. Even at 4%, this is a much more likely scenario than running out of money.

Keith


Keith,

I assume that this is for 1 person, correct? How about a couple Male - 62 years old, Female 54 years old ? ........What would the WR be?


I do not know.

But, here's what I suggest. Go to http://www.immediateannuities.com and look at the options.

A few months ago the payout for a 65-year old was 4% for an inflation protected annuity. I would say the annuity payout is a good approximation to a safe WR.

The difference is of course that the annuity guarantees payments while you live and nothing left after you die. The SWR comes with the possibility of running out of money but also of having a sum left over.

By the way, if you are looking at having some money assured as a legacy, consider buying an SPIA and then term life insurance separately. This is a good alternative to rolling the dice with your own SWR.

And, by the way, the SWR should include your fund and advisor expenses. Your portfolio has an ER of 0.6% and your advisor an AUM 0f 0.72%? Well, 4.0-0.6-0.72 = 2.68% :shock:

That SPIA is looking pretty darn good, I think.

Keith
Déjà Vu is not a prediction
umfundi
 
Posts: 3361
Joined: Tue Jun 07, 2011 6:26 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Valdeselad » Thu Jul 18, 2013 5:19 pm

umfundi wrote:
I think you should also consider failure on the upside: You will live more frugally than you need to, and die with plenty of resources left. Even at 4%, this is a much more likely scenario than running out of money.

Keith


Keith - Thanks for the comment...this was part of what I was getting at in the original post. While you certainly don't want to run out of money, you don't want to be so conservative that you get to the end and realize your estimates were very far off from reality.

I still maintain that the best guide for when you can hang it up is a strong understanding of your expenses, and this is why I diligently track them - month after month, year after year. One reason is to certainly understand how much I spend and what I spend it on. But the other benefit is somewhat subconsciously understanding my spending patterns and behavior. After years of gathering the data, I am very comfortable knowing my patterns of monthly spending that are required to (1) minimally get by, (2) live normally/comfortably or what I would call a "regular" spending month, or (3) buy bigger ticket items, take vacations, eat out more, etc.

This continual self-education of my spending behavior is a key component, I believe, of when I will be comfortable saying that I'm ready to retire and therefore what SWR will be appropriate to make that happen.
Valdeselad
 
Posts: 112
Joined: Mon Apr 26, 2010 2:42 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Cut-Throat » Thu Jul 18, 2013 5:21 pm

umfundi wrote:I do not know.

But, here's what I suggest. Go to http://www.immediateannuities.com and look at the options.

A few months ago the payout for a 65-year old was 4% for an inflation protected annuity. I would say the annuity payout is a good approximation to a safe WR.

The difference is of course that the annuity guarantees payments while you live and nothing left after you die. The SWR comes with the possibility of running out of money but also of having a sum left over.

By the way, if you are looking at having some money assured as a legacy, consider buying an SPIA and then term life insurance separately. This is a good alternative to rolling the dice with your own SWR.

And, by the way, the SWR should include your fund and advisor expenses. Your portfolio has an ER of 0.6% and your advisor an AUM 0f 0.72%? Well, 4.0-0.6-0.72 = 2.68% :shock:

That SPIA is looking pretty darn good, I think.

Keith


I did not see an inflation adjusted annuity in the link that you posted.......Is there one and I missed it?
User avatar
Cut-Throat
 
Posts: 1975
Joined: Sun Oct 17, 2010 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby ObliviousInvestor » Thu Jul 18, 2013 5:29 pm

Cut-Throat wrote:
umfundi wrote:I do not know.

But, here's what I suggest. Go to http://www.immediateannuities.com and look at the options.

A few months ago the payout for a 65-year old was 4% for an inflation protected annuity. I would say the annuity payout is a good approximation to a safe WR.

The difference is of course that the annuity guarantees payments while you live and nothing left after you die. The SWR comes with the possibility of running out of money but also of having a sum left over.

By the way, if you are looking at having some money assured as a legacy, consider buying an SPIA and then term life insurance separately. This is a good alternative to rolling the dice with your own SWR.

And, by the way, the SWR should include your fund and advisor expenses. Your portfolio has an ER of 0.6% and your advisor an AUM 0f 0.72%? Well, 4.0-0.6-0.72 = 2.68% :shock:

That SPIA is looking pretty darn good, I think.

Keith


I did not see an inflation adjusted annuity in the link that you posted.......Is there one and I missed it?


Immediateannuities.com does not offer inflation-adjusted annuities. The only method I'm aware of for getting online inflation-adjusted annuity quotes is the Income Solutions platform (accessible via Vanguard as well).

Click the "get a quote" link on this page to access it:
https://investor.vanguard.com/what-we-o ... eed-income
Mike Piper, author/blogger
User avatar
ObliviousInvestor
 
Posts: 2218
Joined: Tue Mar 17, 2009 10:32 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby umfundi » Thu Jul 18, 2013 5:30 pm

Cut-Throat wrote:I did not see an inflation adjusted annuity in the link that you posted.......Is there one and I missed it?


Yes, I probably screwed up again. The 4% comes from a paper by Wade Pfau from a few months ago. I found it striking that a DIY (inflation-indexed) SWR was the same as the real annuity payout.

Maybe others can provide sources for real annuity payouts? I recall Nisiprius noted that one such site has recently disappeared.

Keith

Edit: Mike Piper to my rescue! :wink:
Déjà Vu is not a prediction
umfundi
 
Posts: 3361
Joined: Tue Jun 07, 2011 6:26 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Cut-Throat » Thu Jul 18, 2013 5:41 pm

umfundi wrote:Yes, I probably screwed up again. The 4% comes from a paper by Wade Pfau from a few months ago. I found it striking that a DIY (inflation-indexed) SWR was the same as the real annuity payout.

Maybe others can provide sources for real annuity payouts? I recall Nisiprius noted that one such site has recently disappeared.

Keith

Edit: Mike Piper to my rescue! :wink:


In the past I have ran quotes on a inflation adjusted annuity for a couple and the numbers that I came up with were pretty paltry. The numbers look great for a non inflation adjusted amount for a single male that is 80 years old, but decline dramatically for most of our 'real life' situations.
User avatar
Cut-Throat
 
Posts: 1975
Joined: Sun Oct 17, 2010 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby ourbrooks » Thu Jul 18, 2013 7:17 pm

One of Wade Pfau's studies suggests that a combination of stocks and fixed payment SPIAs works pretty well, especially for the next few years, when interest rates may be below historical averages. The underlying mechanism is that because of the funds coming in from the SPIA, retirees can hold off longer before selling significant quantities of stocks, which decreases the damage done by a bad market earlier in retirement. In later retirement, the stocks are sold to provide inflation compensation for the SPIAs. No bonds wanted/needed.

There are really two kinds of events that limit the safe withdrawal rate. One is an extended period of poor market returns early in retirement; the other is an extended period of low real bond returns, like we're having now. Until recently, most people overlooked the low bond returns problem but it's that problem that causes portfolios with high percentage of bonds to do poorly.

There's also the possibility of an SPIA ladder in which later SPIAs are purchased to provide inflation compensation for the ones purchased earlier. There's still the problem of how to invest the money that's set aside for the later SPIAs, but it turns out that you don't even need to keep up with inflation, since the older you get the cheaper an annuity gets.
ourbrooks
 
Posts: 1135
Joined: Fri Nov 13, 2009 5:56 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Cut-Throat » Fri Jul 19, 2013 8:14 am

ourbrooks wrote:There are really two kinds of events that limit the safe withdrawal rate. One is an extended period of poor market returns early in retirement; the other is an extended period of low real bond returns, like we're having now. Until recently, most people overlooked the low bond returns problem but it's that problem that causes portfolios with high percentage of bonds to do poorly.


Do you have a historical example of where low Bond returns produced a 'limited safe withdrawal rate'?
User avatar
Cut-Throat
 
Posts: 1975
Joined: Sun Oct 17, 2010 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby k66 » Fri Jul 19, 2013 10:34 am

Have a peek at this paper. The math is not beginner-level, but the results are useful. It is based neither on direct historical data nor Monte Carlo simulation.

http://qwema.ca/pdf_research/2007SEPT_SustSpending.pdf

Purview of the results in the various Tables might give some idea as to what the associated Risk of Ruin is w.r.t given withdrawal rates. It also considers the case of an infintely-lived portfolio, i.e. a Trust.
User avatar
k66
 
Posts: 119
Joined: Sat Oct 27, 2012 2:36 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby ourbrooks » Fri Jul 19, 2013 11:15 am

Cut-Throat wrote:
ourbrooks wrote:There are really two kinds of events that limit the safe withdrawal rate. One is an extended period of poor market returns early in retirement; the other is an extended period of low real bond returns, like we're having now. Until recently, most people overlooked the low bond returns problem but it's that problem that causes portfolios with high percentage of bonds to do poorly.


Do you have a historical example of where low Bond returns produced a 'limited safe withdrawal rate'?


Pick any 30 year period starting between 1937 and 1955. During all of those years, real bond returns were essentially zero. If you had a 100% bond portfolio, you were, effectively, spending your capital every year. If you withdrew 4%, you were spending 1/25 of your portfolio each year, so there was no way it would last 30 years.
ourbrooks
 
Posts: 1135
Joined: Fri Nov 13, 2009 5:56 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Cut-Throat » Fri Jul 19, 2013 11:55 am

ourbrooks wrote:
Cut-Throat wrote:
ourbrooks wrote:There are really two kinds of events that limit the safe withdrawal rate. One is an extended period of poor market returns early in retirement; the other is an extended period of low real bond returns, like we're having now. Until recently, most people overlooked the low bond returns problem but it's that problem that causes portfolios with high percentage of bonds to do poorly.


Do you have a historical example of where low Bond returns produced a 'limited safe withdrawal rate'?


Pick any 30 year period starting between 1937 and 1955. During all of those years, real bond returns were essentially zero. If you had a 100% bond portfolio, you were, effectively, spending your capital every year. If you withdrew 4%, you were spending 1/25 of your portfolio each year, so there was no way it would last 30 years.


Which is why almost no one advocates a 100% bond Portfolio.
What I don't understand is the 'This time is different' mentality when it comes to SWRs. That maybe 2% is now Safe, and that we are in a 'Worst Case' Period Historically.

In 1937 we also had a similar PE ratio for stocks as we do today. And a 30% Stock/70% Bond Portfolio still had an SWR of 3.5% for a 30 year period, despite real bonds returns being zero.
User avatar
Cut-Throat
 
Posts: 1975
Joined: Sun Oct 17, 2010 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby ourbrooks » Fri Jul 19, 2013 3:51 pm

Well, a lot of people seem to think that "age in bonds" is a financial law. For someone retiring at age 65, that puts their mean bond allocation over a thirty year period at 80%; that's pretty near the allocations that lead to higher portfolio failure. Does it matter that that allocation is reached over time? Well, rodc claims that he gets the same results whether he used the mean allocation or the glidepath. If returns do prove to be lower than the historical average, a lot of these folks are going to be very unhappy.

More generally, the 4% figure is based on a certain conditions such as a particular range of asset allocations and a 30 year time period. A lot of folks are operating near the edge of those conditions or outside of them. Among the people who might be in trouble if stock and bond returns are at the low end of their historical range might be the following:

1. A couple in which the primary wage earner is ten years older than the spouse. They need 40 years of money, not 30.
2. People who own a very high percentage of bonds because bonds are "safer."
3. People who plan on doing a higher withdrawal rate early in retirement.

It wouldn't hurt to get anyone falling into those classes to consider a lower withdrawal rate, at least, for the first five years.
ourbrooks
 
Posts: 1135
Joined: Fri Nov 13, 2009 5:56 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Cut-Throat » Fri Jul 19, 2013 4:04 pm

ourbrooks wrote:Well, a lot of people seem to think that "age in bonds" is a financial law. For someone retiring at age 65, that puts their mean bond allocation over a thirty year period at 80%; that's pretty near the allocations that lead to higher portfolio failure. Does it matter that that allocation is reached over time? Well, rodc claims that he gets the same results whether he used the mean allocation or the glidepath. If returns do prove to be lower than the historical average, a lot of these folks are going to be very unhappy.



'Age in Bonds' is a Rule of Thumb, not something that should be followed to the letter. As an example Vanguard's Target Retirement Income is never more than 70% in Bonds. All of their Target Funds use a Glidepath to this 70% Bond Allocation, but Never more than 70%.

BTW - My personal WR is 3% of Portfolio Balance with an increasing percentage at age 70, 80 etc..... --- Which is a flexible withdrawal rate. I am much more comfortable with a Flexible Withdrawal Rate than trying to swing for the fences with a higher stock allocation.
User avatar
Cut-Throat
 
Posts: 1975
Joined: Sun Oct 17, 2010 10:46 am

Re: Trying to Better Understand Safe Withdrawal Rates

Postby umfundi » Fri Jul 19, 2013 8:23 pm

Food for thought.

http://wpfau.blogspot.ca/2013/07/the-bo ... ource.html


I find these figures very thought provoking. There is no doubt the relation between years and withdrawal rate is an hyperbola.

Unknowable future conditions will move the curves up or down, left or right some, but the relative implications of say 5% vs 4% vs 3% will stay the same. (Because, for each of us, the future will be what it will be.)

Keith
Déjà Vu is not a prediction
umfundi
 
Posts: 3361
Joined: Tue Jun 07, 2011 6:26 pm

Re: Trying to Better Understand Safe Withdrawal Rates

Postby Cut-Throat » Fri Jul 19, 2013 9:08 pm

umfundi wrote:Food for thought.

http://wpfau.blogspot.ca/2013/07/the-bo ... ource.html


I find these figures very thought provoking. There is no doubt the relation between years and withdrawal rate is an hyperbola.

Unknowable future conditions will move the curves up or down, left or right some, but the relative implications of say 5% vs 4% vs 3% will stay the same. (Because, for each of us, the future will be what it will be.)

Keith


As you said, the curves are probably accurate. Whether they happen to be 3,4, or 5% is a complete unknown. But that never stops anyone from trying to come up with a formula that predicts the future. :happy
User avatar
Cut-Throat
 
Posts: 1975
Joined: Sun Oct 17, 2010 10:46 am


Return to Investing - Theory, News & General

Who is online

Users browsing this forum: ababs, claver, dr.j, emoore, gwrvmd, hornet96, Leeraar, MnD, munemaker, SMA64, sonny1022, spectec, vturn and 68 guests