Projecting “Increasing Withdrawal Rates” and not SWR

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Bob
Posts: 213
Joined: Mon Feb 26, 2007 4:15 pm

Projecting “Increasing Withdrawal Rates” and not SWR

Post by Bob »

I wonder if anyone has a POV on the reality of planning for constant, inflation-adjusted withdrawals from savings during retirement? For instance, is planning for flat sustainable withdrawal rates (SWR) real, even if adjusted for inflation.

My DW and I are retiring soon and so I did a detailed, bottom-up analysis of all our expenses, income and required withdrawals from savings for the next 15 years. It is a conservative forecast which assumes today’s low rates interest continue for another 5 years, recent COLA rate increases in Social Security and in Medicare continue, our historical costs increases in housing (i.e. property taxes, insurance, etc. ) and OOP healthcare costs rising at 1.5X the CPI. The result is that our annual withdrawals from savings rise over time at about 2X the rate of inflation, not at the rate of inflation.

Now I totally buy Yogi Berra’s warning that “It's tough to make predictions, especially about the future” so my own projections are just one guess at a scenario. But I wonder what other people expect for their own annual withdrawals from retirement savings and think are the implications?
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Rodc »

I guess you need to define "planning"

I suspect few people have constant real dollar desires. Or needs either. But then again life rarely goes according to plans or at least according to detailed plans. If you have a good proportion of desired spending in discretionary you can and likely will have a fair amount of variability vs plan.

So, how many significant digits should your plan have?

A "plan" in pencil that is just a rough gauge of desires and income that is constant might work for many.

A modest modification (say you retire at 63 and plan to take SS at 67, so you account for 4 years of covering not having SS in your otherwise flat plan) might be good enough for many more.

Just depends.

I'd say it is not a good actual action plan, just a rough plan to gauge where you are and depending on your situation might be of some value.
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.
User avatar
cheese_breath
Posts: 11769
Joined: Wed Sep 14, 2011 7:08 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by cheese_breath »

What you are talking about sounds a lot like my Excel financial projection spreadsheet. I've mentioned it several times in other threads, but in case you haven't seen them...

The spreadsheet projects on a monthly basis my expected expenses, income, required withdrawals, and resulting portfolio value. It groups expenses into various categories instead of trying to predict at too great a detail and includes variables for assumed inflation, COLA, and investments growth over my and wife's expected lifetimes. These may be modified to perform 'what if' scenarios. I developed it and performed my first projection when I retired in 1997 and have been tracking to that projection ever since.
The surest way to know the future is when it becomes the past.
User avatar
obgyn65
Posts: 770
Joined: Sun May 06, 2012 9:11 am

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by obgyn65 »

Increasing withdrawing rates don't make sense to me, sorry. Either NW withdrawal rates (e.g constant, variable, etc) are safe or they are not. If they are not, expenses must be curtailed or you should work a bit longer. My 2 cents only. FYI, my planned safe withdrawal rate is constant.
"The two most important days in someone's life are the day that they are born and the day they discover why." -John Maxwell
Longdog
Posts: 2173
Joined: Sun Feb 09, 2014 5:56 pm
Location: Philadelphia

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Longdog »

obgyn65 wrote:Increasing withdrawing rates don't make sense to me, sorry. Either NW withdrawal rates (e.g constant, variable, etc) are safe or they are not. If they are not, expenses must be curtailed or you should work a bit longer. My 2 cents only. FYI, my planned safe withdrawal rate is constant.
A constant dollar amount or a constant percentage of assets?
Steve
User avatar
midareff
Posts: 7711
Joined: Mon Nov 29, 2010 9:43 am
Location: Biscayne Bay, South Florida

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by midareff »

obgyn65 wrote:Increasing withdrawing rates don't make sense to me, sorry. Either NW withdrawal rates (e.g constant, variable, etc) are safe or they are not. If they are not, expenses must be curtailed or you should work a bit longer. My 2 cents only. FYI, my planned safe withdrawal rate is constant.
+1 Fixed withdrawal rate as a % and the funds available for withdrawal will vary with the value of the portfolio. Add to that fixed rate a ceiling and floor and that's my plan. I'm at floor right now at 3.0% and still accumulating excess cash which I return to the portfolio every few months. Computationally a 40% decline of equities would raise my rate 3.7% and I might think twice about large expenses, a 60% decline would bring it to 4.2% which for a several year sustained decline does not overly effect my behavior as long as withdrawals are from price stable bond funds rather than equities.

I understand your concern is costs of items outrunning the CPI-U indexed income streams.

FWIW, I believed in accumulating enough to not have to sweat lots of things. The run up of the last few years has pushed my safety margin past my expectation level but the basic premises are intact. If you are running numbers through your head in the middle of the night you haven't provided an adequate safety margin. If you don't see a safety margin when liberally accounting for the unknown your choices are save more and work longer. Once you retire the value of your human capital will diminish to zero quickly. Being in my 70's and bagging groceries for movie money is not on the agenda. If you find a conservative fixed WR has produced too much extra money there is always a new S Class at the MB dealer.
User avatar
Bustoff
Posts: 2033
Joined: Sat Mar 03, 2012 5:45 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Bustoff »

Bob wrote: I wonder if anyone has a POV on the reality of planning for constant, inflation-adjusted withdrawals from savings during retirement?

Prior to retirement, I invested many hours into planning. It was a waste of time for a couple of reasons. First, I retired sooner than planned. Second, since retiring much of our life has changed, both in terms of goals and assumptions.
Bob wrote:I wonder what other people expect for their own annual withdrawals from retirement savings and think are the implications
The only thing I can control in retirement is my spending which is based on, hopefully, "secure" sources of income.
The remaining risk portfolio giveth and taketh away, whereby spending is adjusted accordingly.

You may find this Vanguard article helpful:
The 4% spending rule, 20 years later

Best of luck with your retirement plans!

P.S. Planing for 35X annual residual expenses might be a consideration.
Last edited by Bustoff on Sat May 02, 2015 8:43 am, edited 1 time in total.
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by BahamaMan »

Having worked and invested for over 30 years, I saved all I could and lived a 'balanced' life taking Vacations and enjoying life within a Budget. That was a "Saving and Investing" Budget. Now that I am retired I have a "Spending" Budget. I am no longer concerned with Saving and Investing and I spend whatever my "Spending Budget" says I can. Economics was defined to me as "Having Unlimited Wants, with Limited Finances".... I would have no trouble spending $100 Million if I had it. If I could afford a Private Jet, I'd buy one. But since I can't I 'Put up' with Commercial Air Travel.

I do this by employing VPW (Variable Percentage Withdrawal). Which is an ever increasing percentage of your Portfolio Balance. My first year of Retirement Had me spending twice of what I was spending before I was retired. Of course I now spend 3 months of Winter in the Bahamas and take a lot more fishing trips. It is a safer withdrawal method than a Fixed Amount withdrawal, as it will reduce withdrawals if your portfolio balance declines. My Worstcase Historical Downside Spending Amount is about a 15% decline for my A.A. (Which includes my S.S., Withdrawal Amount and Small Pension). Read through this and download the software and backtest it, and see if it suits you. It is a perfect withdrawal tool and method for me. http://www.bogleheads.org/wiki/Variable ... withdrawal
Last edited by BahamaMan on Sat Feb 14, 2015 9:23 am, edited 1 time in total.
Dandy
Posts: 6701
Joined: Sun Apr 25, 2010 7:42 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Dandy »

I don't think you can have a confident plan for retirement withdrawals that will be real for much more than a few years. It is great to plan and to get a good sense of likely outcomes for asset growth, withdrawals, inflation etc. Based on that information you can set an initial withdrawal amount.

Every few years you need to check on your health, expenses and assets to determine what adjustments to your withdrawal plan, expenses or investments needs to be made. There are just too many variables that can impact your expenses and assets to set a course and expect not to alter it.
User avatar
Sheepdog
Posts: 5783
Joined: Tue Feb 27, 2007 2:05 pm
Location: Indiana, retired 1998 at age 65

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Sheepdog »

Repeating my message ad nauseam....It does not make sense to me to set a beginning withdrawal figure and increase that automatically by that year's inflation each year regardless of investment performance and/or needs. Instead, a planned average percentage withdrawal rate based on the previous year ending investment balance works. If I have more, I can spend more. If I have less, I must try to spend less. In my case, my annual withdrawal rate was to be an average of 4.5%. It was to be an average percentage because expenses are not constant year to year. Actually, my annual withdrawals ranged from 3.11% to 7.52% over 16 years, but the average is 4.56% thru 2014, very close to the plan. And, my total investment has grown.

Revised to add this:
Bob, you stated " our annual withdrawals from savings rise over time at about 2X the rate of inflation, not at the rate of inflation. ' Regardless of the calculations, I don't know why you would want to do that. Our expenses in our 16 retirement years, with the same standard of living (new cars in certain years, regular nice vacations & cruises, house maintenance, medical and dental, etc) our expense history was, as follows: Our 5 year average spending in the 1999 thru 2003 period was $57,850; 2003 thru 2007 was $57,828; 2007 thru 2011 was $66,406; and the 2011 thru 2014 was $68,495. (Yes, I did overlap one year in each calculation.) A very rough calculation (you may be more accurate) shows that my spending inflation averaged about the inflation rate. If you were to compare just our spending in 1999 ($68,648) with our spending in 2014 ($70,001), you could say my inflation was negligible, but that is not accurate, but it may be interesting to you.
Unless you try to do something beyond what you have already mastered you will never grow. (Ralph Waldo Emerson)
Topic Author
Bob
Posts: 213
Joined: Mon Feb 26, 2007 4:15 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Bob »

Lots of good points and very much appreciate hearing about other peoples' POVs and experience. Thanks.

I totally agree there is a major difference between doing "planning" and actually "managing" ones spending and withdrawals. For managing I expect we will use concepts like VPW to insure we are always playing it safe.

My observation that my projections of our un-managed, likely natural growth in retirement withdrawals may increase at more than the rate of CPI-U inflation was what I was interested in comparing to others thoughts. (Recognizing that the past does not predict the future, but is only a perspective on what might be.)

As others have suggested for their spending, our own spending (excluding income taxes) over the last 10 years has varied year by year but the overall trend has been that it has increased at a CAGR of 4.51% which is above the official U.S. CPI-U.

Image

And so, I was not surprised when my Excel modeling of what our future financials might look like (unmanaged) showed a similar personal rate of inflation above what the official and "experts" numbers suggest.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Rodc »

... instead of trying to predict at too great a detail...
If you are making monthly calculations out your years of retirement I suggest it is possible you have already gone over into the abyss of "too great a detail". :happy
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.
ourbrooks
Posts: 1575
Joined: Fri Nov 13, 2009 3:56 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by ourbrooks »

I think it's perfectly okay to do a detailed plan, down to the monthly level in retirement. Then, do a couple more detailed plans, with different assumptions. This will let you spot at least the major factors regarding your financial situation.

In the OP's case, the detailed planning has worked perfectly; it has allowed the OP to spot a potential retirement spending problem, increasing real retirement spending. I don't think I've ever seen that come up in discussions of retirement spending; they also seem to assume declining spending. If the OP had just done rough planning, following the standard advice, he would not have found out that his situation might be different. Three cheers for detailed planning.

The goal of any planning ought to be timely, appropriate action. The next step for the OP is to find the cause(s) of the increasing spending. Most of the writing on spending in retirement assumes declining spending; why is the OP's situation different? Is housing a primary factor? If so, the OP might start considering steps to reduce those costs in retirement, such as relocating to an area with lower real estate prices, taxes and insurance costs. etc. Maybe it's just that the OP has been taking better and better vacations. Maybe, and I hate to think about this, the OP has just been more realistic than the rest of us about retirement health care costs. Find out why things are different and see if anything can be done about it.

Alas, it might still be the case that any acceptable plan includes rising spending. Maybe the OP is already living in the place of his dreams, in the house of his dreams, and moving elsewhere would be a terrible disappointment. That's point to explore withdrawal plans which are based on rising real spending. Part of any such plan ought to include some thinking about contingencies, what to do if the money does run short.

[BTW, VPW or constant percentage withdrawal or any other variable scheme probably won't help if the problem is that the OP's core real spending is rising. You can't tell your house insurance company that you're reducing your payments this year because the market tanked.]
Topic Author
Bob
Posts: 213
Joined: Mon Feb 26, 2007 4:15 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Bob »

Just to clarify a few items:

Because the DW and I have always lived below our means and saved a large portion of our incomes for the last 40 years we are in a lucky position. According to most all the models and rules of thumb, they predict we need not worry about running out of money even if one of us manages to live past 100. But all those models and rules of thumb make board, simplifying assumptions so I want to try to understand where and what might be wise to plan for as "safety margins" or contingencies. My objective was not just to over-analyze things nor to think I can predict the future. It is as 'ourbrooks' nicely put it: " ...to spot [any likely] potential retirement spending problem [gotchas] ..." that I can then make plans to deal with.

For instance, I think the studies asuggesting people will spend in retirement are very misleading for a lot of people. It does appear true that in the past for the U.S. retired population when you look at the consolidated numbers. But I think that is because a large proportion of the retired U.S. population lives off essentially only Social Security income (23% of married couples, 46% of singles) and is forced to cut their spending as Social Security COLAs have not really kept up with inflation. It is not because that 23%/46% would do so if not forced to do so, is my guess, and that group impacts the consolidated numbers.

For us -- or actually for my annual model "what-if" scenario -- the most important item that drives our increased spending and withdrawals over time is what I have projected for our healthcare costs which initially skyrocket when we shift from great employer-paid benefits to a matrix of plans to try to keep similar coverage as we get older and actually will need it (i.e.: Medicare + Plan F Medigap + Dental + Vision + OOP, since no company provided healthcare insurance when we retire). I have also assumed that healthcare cost inflation will continue to outpace the CPI-U by 2X (as it did 1999 to 2009) even though there has been a slow-down in total healthcare cost increases in the last couple years. While total costs are slowing OUR costs for co-pays and OOPs are going up quickly and a lot, which I expect to continue. So for us, I wanted to be conservative and so I assumed we will actually see costs for healthcare continue to rise faster than overall inflation.
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by BahamaMan »

ourbrooks wrote: [BTW, VPW or constant percentage withdrawal or any other variable scheme probably won't help if the problem is that the OP's core real spending is rising. You can't tell your house insurance company that you're reducing your payments this year because the market tanked.]
If your base expenses are that 'tight' I would not recommend retiring at all.... Or you need to cut your standard of living so that your Base expenses are less than half of your initial withdrawal..... Any withdrawal scheme won't help you, especially the fixed withdrawal ones!
User avatar
obgyn65
Posts: 770
Joined: Sun May 06, 2012 9:11 am

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by obgyn65 »

A constant dollar amount but things may change as my planning horizon is 45 years from now, when I reach 95.
SteveM wrote:
obgyn65 wrote:Increasing withdrawing rates don't make sense to me, sorry. Either NW withdrawal rates (e.g constant, variable, etc) are safe or they are not. If they are not, expenses must be curtailed or you should work a bit longer. My 2 cents only. FYI, my planned safe withdrawal rate is constant.
A constant dollar amount or a constant percentage of assets?
"The two most important days in someone's life are the day that they are born and the day they discover why." -John Maxwell
scifilover
Posts: 530
Joined: Sun Apr 14, 2013 12:56 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by scifilover »

As a person who used to try to do this stuff for a big company, I'd like to remind planners about some of the problems in long term projections.

When you do a spreadsheet, you make a number of assumptions for each year of the plan. These might include various inflation rates for items in the plan, investment return rates for different classes of investments, etc. What most forget to do is to try to determine the probability of accuracy for each of the assumptions. When you make an assumption, you are making a guess about the future. It may or may not happen. So, if in a given year of your spreadsheet, there are 15 assumptions, there are 15 places to be right or wrong. So let's say that there is a 90% probability of accuracy in each of your 15 assumptions. Then, for that given year of your plan the over all probability of your plan coming to pass is the product of the multiplication of all the assumptions or .9 x .9 x .9 and so on 15 times. You must then multiply this aggregate probability times the aggregate probability for each of the 15 years of the projection to determine the probability your plan will work.

The end result is that there is no accuracy in long term projections. Even if you use 95% accuracy for assumptions, the long term accuracy will be close to zero.

So, the value in doing long term planning via spreadsheet is to help you think about the future and consider by varying the assumptions, what the range of outcomes might be.

When I retired in 2001, I did a lot of projections. There was no way to project what has happened to interest rates over that time. My carefully built bond ladder has been trashed, indeed, my fixed income has dwindled to a fraction of what I used to enjoy.

This is why planning tools that use Monte Carlo simulations have more value in helping think about the future in terms of confidence then your own spreadsheet model.
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by BahamaMan »

scifilover wrote:The end result is that there is no accuracy in long term projections. Even if you use 95% accuracy for assumptions, the long term accuracy will be close to zero.

So, the value in doing long term planning via spreadsheet is to help you think about the future and consider by varying the assumptions, what the range of outcomes might be.

When I retired in 2001, I did a lot of projections. There was no way to project what has happened to interest rates over that time. My carefully built bond ladder has been trashed, indeed, my fixed income has dwindled to a fraction of what I used to enjoy.

This is why planning tools that use Monte Carlo simulations have more value in helping think about the future in terms of confidence then your own spreadsheet model.
Exactly! When you are retired, you should have a plan based on the Historical Record and then react accordingly to Real Time conditions. Your Budget should drive your Expenditures. Not the other way around!
Last edited by BahamaMan on Sun Feb 15, 2015 9:46 am, edited 1 time in total.
Topic Author
Bob
Posts: 213
Joined: Mon Feb 26, 2007 4:15 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Bob »

I agree with Scifilover about not expecting any given scenario to be a reliable prediction of the future. And I do understand the power and limitations of Monte Carlo as well as backtesting, etc. They all help me to better understand what are the biggest income and cost areas that when they change due to unpredictable reasons, will drive eventual outcomes. And then my goal will be to find products or ways to mitigate their potential impact.

While I may find it too expensive or impractical, I just want to see if I can minimize the magnitude of a worst case sceanrio outcome. Kind of a minimax strategy.

So far, the major drivers with the greatest variability we cannot likely control and impact on long -term outcomes seem to be:
- healthcare costs (as well as our personal health)
- interest rates on our bond investments
- future tax rates (on Social Security, on Munis, etc.)

Anything that I should add, or reorder in importance?
Leeraar
Posts: 4109
Joined: Tue Dec 10, 2013 7:41 pm
Location: Nowhere

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Leeraar »

I think it is important to do cash flow projections at least through age 70, when you will have claimed SS and begun RMDs. There are many potential variations during the decade 60 - 70: Medical care to age 65, medical care after age 65, claiming SS or not for two peple (presuming you have a partner), getting the last kid through college, paying off the mortgage or purchasing a retirement residence, etc.

We are delaying SS and yet our withdrawal rate is 2.5%, so we clearly do not have a problem. But, we have some large discretionary items, such as gifting to our kids to fund their Roth IRAs until they are 30. And, we need to spend $40,000 to replace our deck and driveway and redo the landscaping. We also have one more car than we really need.

I have not bothered to put inflation into my projections. rather, I have been concentrating on annuitizing (through SPIAs and SS when we claim it) our base required income. Our spending needs when I turn 70, in five years? I guess we'll know when we get there.

L.
You can get what you want, or you can just get old. (Billy Joel, "Vienna")
Leeraar
Posts: 4109
Joined: Tue Dec 10, 2013 7:41 pm
Location: Nowhere

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Leeraar »

Bob wrote:Anything that I should add, or reorder in importance?
Look at your expenses.

Transportation. For us, insurance is far greater than the lease payments on the vehicles.

Housing, including property taxes. Are you prepared to move to a lower cost location?

Cell phones and cable TV and Netflix, etc. Could be a large chunk.

L.
You can get what you want, or you can just get old. (Billy Joel, "Vienna")
heyyou
Posts: 4452
Joined: Tue Feb 20, 2007 3:58 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by heyyou »

Your Budget should drive your Expenditures.
Well said by BahamaMan. DW and I expect to adapt our spending to our income.
Money solves some problems by having either a low withdrawal rate or discretionary spending that can be easily curtailed.
Striving to have planned for most scenarios could be a retirement hobby, but to some it appears to be looking for certainty in an uncertain future. You will get the returns that are specific to your precise time in the markets, not some average of all of the other periods. Perhaps monitoring as you go, is a good path.
User avatar
grabiner
Advisory Board
Posts: 35265
Joined: Tue Feb 20, 2007 10:58 pm
Location: Columbia, MD

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by grabiner »

Bob wrote:My DW and I are retiring soon and so I did a detailed, bottom-up analysis of all our expenses, income and required withdrawals from savings for the next 15 years. It is a conservative forecast which assumes today’s low rates interest continue for another 5 years, recent COLA rate increases in Social Security and in Medicare continue, our historical costs increases in housing (i.e. property taxes, insurance, etc. ) and OOP healthcare costs rising at 1.5X the CPI. The result is that our annual withdrawals from savings rise over time at about 2X the rate of inflation, not at the rate of inflation.
While your withdrawals may increase faster than inflation, they shouldn't increase at a fixed multiple of the rate; if they increase by 3% with 2% inflation, they are unlikely to increase by 15% with 10% inflation (particularly if your state has protections against rapidly rising property taxes for retired homeowners), or by zero if there is no inflation. It would be reasonable to project withdrawals which grow 1% faster than inflation.
Wiki David Grabiner
User avatar
cheese_breath
Posts: 11769
Joined: Wed Sep 14, 2011 7:08 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by cheese_breath »

Whatever tools or techniques you use to do your projections it's important to understand that the results aren't static. As time passes and you learn more you can better refine your numbers and assumptions. I've refined assumptions in my Excel spreadsheet several times over the past 18 years. And at the end of every month I replace my projected expenses for that month with actual data. This gives me a little better idea of what the next year's expense might be and rolls through the years all the way to the end.
The surest way to know the future is when it becomes the past.
ourbrooks
Posts: 1575
Joined: Fri Nov 13, 2009 3:56 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by ourbrooks »

1. Consider some scenarios which replace bonds with SPIAs. Wade Pfau argues that in low interest rate environments, this will yield better results.

2. What about irregular large expenses, such as new cars or new roofs? A common way to model these is as annual "savings." The problem is that if you save over a long period of time you are (a) depriving yourself of the benefits of tax shielded growth and (b) depriving yourself of the higher return rates if you left the money in your portfolio. Instead, model them as lump some withdrawals.

3. Is your home suitable for someone who is mobility impaired or who can no longer drive. There are lots of otherwise healthy people who can no longer make it up and down stairs or who can no longer drive at night. Moving is harder the older you get and moving into a retirement facility because you can no longer live at home is particularly hard. In your financial modeling, consider the costs of making your current home more accessible or of moving if your home is located where you must always drive to get things done.
ralph124cf
Posts: 2985
Joined: Tue Apr 01, 2014 11:41 am

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by ralph124cf »

Sheepdog wrote:Repeating my message ad nauseam....It does not make sense to me to set a beginning withdrawal figure and increase that automatically by that year's inflation each year regardless of investment performance and/or needs. Instead, a planned average percentage withdrawal rate based on the previous year ending investment balance works. If I have more, I can spend more. If I have less, I must try to spend less. In my case, my annual withdrawal rate was to be an average of 4.5%. It was to be an average percentage because expenses are not constant year to year. Actually, my annual withdrawals ranged from 3.11% to 7.52% over 16 years, but the average is 4.56% thru 2014, very close to the plan. And, my total investment has grown.

Revised to add this:
Bob, you stated " our annual withdrawals from savings rise over time at about 2X the rate of inflation, not at the rate of inflation. ' Regardless of the calculations, I don't know why you would want to do that. Our expenses in our 16 retirement years, with the same standard of living (new cars in certain years, regular nice vacations & cruises, house maintenance, medical and dental, etc) our expense history was, as follows: Our 5 year average spending in the 1999 thru 2003 period was $57,850; 2003 thru 2007 was $57,828; 2007 thru 2011 was $66,406; and the 2011 thru 2014 was $68,495. (Yes, I did overlap one year in each calculation.) A very rough calculation (you may be more accurate) shows that my spending inflation averaged about the inflation rate. If you were to compare just our spending in 1999 ($68,648) with our spending in 2014 ($70,001), you could say my inflation was negligible, but that is not accurate, but it may be interesting to you.
I have no problems with the variable spending, if you can afford it. If the withdrawals are from taxable savings, or a Roth, then again no problem.

The problem that can arise is if the withdrawals are from tax sheltered accounts, then the changing marginal tax rates of high versus low withdrawals may well cost extra taxes. If you can keep withdrawals (and thus taxable income) stable, then you may pay less income tax over several years. You can accommodate variable spending by keeping a bigger buffer in either a taxable account or a Roth, and keeping your tax deferred withdrawals at an optimum average amount considering taxes. If this average withdrawal is higher than you need to spend, just convert it to a Roth where it will continue to grow tax free, and be available for tax free withdrawal when your spending spikes.

Ralph
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by BahamaMan »

ralph124cf wrote: I have no problems with the variable spending, if you can afford it. If the withdrawals are from taxable savings, or a Roth, then again no problem.

The problem that can arise is if the withdrawals are from tax sheltered accounts, then the changing marginal tax rates of high versus low withdrawals may well cost extra taxes. If you can keep withdrawals (and thus taxable income) stable, then you may pay less income tax over several years. You can accommodate variable spending by keeping a bigger buffer in either a taxable account or a Roth, and keeping your tax deferred withdrawals at an optimum average amount considering taxes. If this average withdrawal is higher than you need to spend, just convert it to a Roth where it will continue to grow tax free, and be available for tax free withdrawal when your spending spikes.

Ralph

This is a Problem with Withdrawals from a Tax sheltered account, not Variable Withdrawals.
The problem would get even worse with a Fixed SWR plus inflation adjustment. IOW - The Market Tanks, and you not only have to withdraw your 4% SWR plus inflation in a down market, but you have to pay the taxes as well! At least with the Variable Withdrawal Method you take less and your tax bill is less! You can mitigate the volatility with a conservative Asset Allocation and Delaying S.S. to age 70.

As you indicated, I plan to mitigate this situation, by converting as much Regular IRA money into Roth Before my S.S. kicks in, so that I can 'Buffer' the tax consequences as much as possible.
pkcrafter
Posts: 15461
Joined: Sun Mar 04, 2007 11:19 am
Location: CA
Contact:

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by pkcrafter »

Bob wrote:I wonder if anyone has a POV on the reality of planning for constant, inflation-adjusted withdrawals from savings during retirement? For instance, is planning for flat sustainable withdrawal rates (SWR) real, even if adjusted for inflation.

My DW and I are retiring soon and so I did a detailed, bottom-up analysis of all our expenses, income and required withdrawals from savings for the next 15 years. It is a conservative forecast which assumes today’s low rates interest continue for another 5 years, recent COLA rate increases in Social Security and in Medicare continue, our historical costs increases in housing (i.e. property taxes, insurance, etc. ) and OOP healthcare costs rising at 1.5X the CPI. The result is that our annual withdrawals from savings rise over time at about 2X the rate of inflation, not at the rate of inflation.

Now I totally buy Yogi Berra’s warning that “It's tough to make predictions, especially about the future” so my own projections are just one guess at a scenario. But I wonder what other people expect for their own annual withdrawals from retirement savings and think are the implications?
Bob, what is driving up your projected costs--mainly health care? What increase in withdrawals are you projecting.

By the way, I suspect your calculations are probably more realistic than just using the "reported" inflation rate.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.
Topic Author
Bob
Posts: 213
Joined: Mon Feb 26, 2007 4:15 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Bob »

Thank you to all for the excellent ideas and counsel. It is great to obtain outside perspectives that I can use to stretch my thinking.

To answer 'pkcrafter's' question, yes, healthcare is a major driver to above CPI-U inflation in withdrawals is as we move from employer-based heath benefits to Medicare and the myriad of other insurance types/plans that one might buy to equal what we have now (e.g.: costs for Medigap, Vision and Dental plans instead of self-insuring). Really makes me appreciate how lucky we have been mohave those benefits :happy
User avatar
Sheepdog
Posts: 5783
Joined: Tue Feb 27, 2007 2:05 pm
Location: Indiana, retired 1998 at age 65

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by Sheepdog »

ralph124cf wrote:
I have no problems with the variable spending, if you can afford it. If the withdrawals are from taxable savings, or a Roth, then again no problem.

The problem that can arise is if the withdrawals are from tax sheltered accounts, then the changing marginal tax rates of high versus low withdrawals may well cost extra taxes. If you can keep withdrawals (and thus taxable income) stable, then you may pay less income tax over several years. You can accommodate variable spending by keeping a bigger buffer in either a taxable account or a Roth, and keeping your tax deferred withdrawals at an optimum average amount considering taxes. If this average withdrawal is higher than you need to spend, just convert it to a Roth where it will continue to grow tax free, and be available for tax free withdrawal when your spending spikes.

Ralph
Ralph,
This is how I worked on that problem for my wife and my retirement funding and reduced taxes considerably.
Before I retired in '98, I developed a plan to reduce my taxable accounts as much as possible in my first years in retirement. In the first 5 years I lived off of only the taxable accounts and SS, plus I bought I Bonds as much as practical (about 20% of my total investments) which, in effect, increased my tax deferred investments as well. In that time, I also converted some of my tIRAs to Roths equal to about 22% of my total investments. The result is that without a lot of taxable accounts and no pension (took mine as a lump sum and converted to a tIRA) I pay little to no taxes...zero in most years since 2002. After SS, I take from our tIRAs until taxes would be due, then take from our Roths, if needed, to accomplish that blessing.
Unless you try to do something beyond what you have already mastered you will never grow. (Ralph Waldo Emerson)
cjking
Posts: 2039
Joined: Mon Jun 30, 2008 4:30 am

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by cjking »

BahamaMan wrote:
ralph124cf wrote: I have no problems with the variable spending, if you can afford it. If the withdrawals are from taxable savings, or a Roth, then again no problem.

The problem that can arise is if the withdrawals are from tax sheltered accounts, then the changing marginal tax rates of high versus low withdrawals may well cost extra taxes. If you can keep withdrawals (and thus taxable income) stable, then you may pay less income tax over several years. You can accommodate variable spending by keeping a bigger buffer in either a taxable account or a Roth, and keeping your tax deferred withdrawals at an optimum average amount considering taxes. If this average withdrawal is higher than you need to spend, just convert it to a Roth where it will continue to grow tax free, and be available for tax free withdrawal when your spending spikes.

Ralph

This is a Problem with Withdrawals from a Tax sheltered account, not Variable Withdrawals.
The problem would get even worse with a Fixed SWR plus inflation adjustment. IOW - The Market Tanks, and you not only have to withdraw your 4% SWR plus inflation in a down market, but you have to pay the taxes as well! At least with the Variable Withdrawal Method you take less and your tax bill is less! You can mitigate the volatility with a conservative Asset Allocation and Delaying S.S. to age 70.

As you indicated, I plan to mitigate this situation, by converting as much Regular IRA money into Roth Before my S.S. kicks in, so that I can 'Buffer' the tax consequences as much as possible.
Not sure that the first paragraph in this response has got to grips with the issue. Let's say that I (not in US) pay 20% tax on income up to 40K, and 40% tax on income above that. Initially all my money is in a tax-deferred account. If I take constant real 40K, I never pay 40% tax. (Presumably inflation increases in income are matched by increases in the tax threshold, if not one would just make insignificant/small adjustments.) If I take variable income averaging 40K, so for example 50K one year and 30K the next, clearly my tax bill will be higher for the same total before-tax income. Not sure how you can say it isn't the variability of withdrawals that is the problem here. (I know that smooth withdrawals causes it's own problems, but that is not the issue being discussed here. I can see potential workarounds to the tax problem of variable withdrawals, but saying potential workarounds exist is not the same as saying the problem doesn't exist.)
bhsince87
Posts: 2914
Joined: Thu Oct 03, 2013 1:08 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by bhsince87 »

I think it's important to consider the possibility that our "personal CPI" might be very different than the government's CPI upon which TIPS, Social Security, annuities, pensions, etc, are based.

For example, housing is over 40% of the official CPI. But if you live in a paid for house, it might be a very small part of your total expenses. Education is 7%, but it's about 0% of my spending. Medical care is 7.5% of the official rate, but I expect it will be much higher than that for my wife and I, at least until we hit medicare eligibility.

Here's a blog post discussing the various factors that go into the official number. http://www.advisorperspectives.com/dsho ... erview.php

So it's possible that SS COLA's, TIPS, etc might not be the best inflation protection for everyone.

I'm trying to manage some of this by using some non-Bogleheadish investments. We heat with oil, and expect to drive more when we retire, so I've tried to hedge that a bit with the Vanguard Energy fund. I also own some Health Care fund to try to offset excessive health care costs.
Time is what we want most, but what we use worst. William Penn
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by BahamaMan »

cjking wrote:Not sure that the first paragraph in this response has got to grips with the issue. Let's say that I (not in US) pay 20% tax on income up to 40K, and 40% tax on income above that. Initially all my money is in a tax-deferred account. If I take constant real 40K, I never pay 40% tax. (Presumably inflation increases in income are matched by increases in the tax threshold, if not one would just make insignificant/small adjustments.) If I take variable income averaging 40K, so for example 50K one year and 30K the next, clearly my tax bill will be higher for the same total before-tax income. Not sure how you can say it isn't the variability of withdrawals that is the problem here. (I know that smooth withdrawals causes it's own problems, but that is not the issue being discussed here. I can see potential workarounds to the tax problem of variable withdrawals, but saying potential workarounds exist is not the same as saying the problem doesn't exist.)
You can set up various 'Straw Mans' for a given situation, and it may be worse one way or another. But that is not the issue. First of all, the variability is not as great as 50K one year and 30K the next. That's a 40% decline! The facts are that you don't know the variability. If the market always increased, there would be none. So, this is a lot of speculation on what will be more tax friendly.

And, I'll say it again....The problem is not with Variable Withdrawals, the problem is with Tax Sheltered Accounts. What happens in the future is an unknown. Is it possible that the sequence of returns is such that Variable withdrawals are taxed higher than Smooth ones....Yes, I am sure that you could create a scenario in which that is so. You could also create a scenario where 100% Stock Allocation is better than a 50/50 Portfolio. It doesn't mean you should do it however.
john94549
Posts: 4638
Joined: Tue Jul 26, 2011 8:50 pm

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by john94549 »

We simplified matters by setting aside enough in IRA CDs to cover the first fifteen years or so. Our ladder throws off a real return, so inflation is not a big issue.
itstoomuch
Posts: 5343
Joined: Mon Dec 15, 2014 11:17 am
Location: midValley OR

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by itstoomuch »

OP: if I understand your question, You want to project future WitdrawalRate to match future expenses. ?

Most difficult but low fee: Proper market time and asset allocation.

DIfficult, learning curve, and Limited: Spread Options, Calls, and Puts in a LEAP. Low fees but you risk premium. Rewards could be very high. Limited time frame 1-5 year's into future.

Relatively easy, medium effort, high fee. Guaranteed withdrawal rates out 30 years (initial age dependent). Lifetime Income from day one or deferred . Step ups. Remainder to heirs. GLWB Variable Annuity or GLWB Fixed(Equity)-Income Annuities. Features and guaranteed benefits are becoming less prevalent in low bond environment, inflation/deflation. You will get the most benefit by REALLY understanding the Annuity types. Not well receive by some BH.

See:
viewtopic.php?f=10&t=158409
viewtopic.php?t=155856&f=10
viewtopic.php?f=11&t=157309

Disclaimer: We purchased both GWLB VA and FIA.
YMMV. I have viewed only two (Vanguard/TransAmerica and Jeffereson) in the last year and they are significantly less featured than the ones we purchased 6 & 3 yrs ago.
Again ,YMMV,
:sharebeer GL
Rev012718; 4 Incm stream buckets: SS+pension; dfr'd GLWB VA & FI anntys, by time & $$ laddered; Discretionary; Rentals. LTCi. Own, not asset. Tax TBT%. Early SS. FundRatio (FR) >1.1 67/70yo
cjking
Posts: 2039
Joined: Mon Jun 30, 2008 4:30 am

Re: Projecting “Increasing Withdrawal Rates” and not SWR

Post by cjking »

BahamaMan wrote:So, this is a lot of speculation on what will be more tax friendly.
Not speculation, arithmetical fact. If total income and total number of years is the same, the best variable income can do is result in the same tax bill as constant income, it can never generate a lower bill. In my example tax system this could happen if the constant income was below the higher tax rate threshold, and the variable income was able to produce the same overall income while never having a single year when it went over the threshold.

Anyway, it doesn't matter, as I think one can work around the problem somewhat by setting up a cash cache :) within the tax-deferred account to park any excess income while awaiting a down year in which it can be used to boost the actual withdrawal up to 40K. The cash cache would be regarded as a reserve of spending money that was separate from the volatile portfolio. With the correct mental accounting one would "take" variable withdrawals from the volatile portfolio, but use the cash cache to smooth taxable withdrawals from the account to a constant 40K a year, thus reducing the tax bill.
Post Reply