The Shortcomings of SWR After Retirement

[Midpack]

Interesting POVs. While I use spreadsheets to analyze, it's easy, I usually learn something worthwhile with each exercise, and it takes a trivial amount of time in the overall scheme of things given what's at stake in $. The OP took less than 45 minutes, and I thought it might interest others, though I did it for myself.

And there's a large playing field between 'multiply by 4% and be done with it' and 'analyzing to the nth degree' - few here probably go to either extreme, especially the latter. I think am somewhere in the middle, simply trying to make good fundamental choices as there are many variables and unknowns as we all know, and some known changes that occur from time to time, like tax code. By all means to each his/her own...

 

[samclem]

And there's a large playing field between 'multiply by 4% and be done with it' and 'analyzing to the nth degree' - few here probably go to either extreme, especially the latter. I think am somewhere in the middle, simply trying to make good fundamental choices as there are many variables and unknowns as we all know, and some known changes that occur from time to time, like tax code. By all means to each his/her own...

I agree--playing with i-orp, doing a spreadsheet, etc, is how I came to appreciate the importance of small differences in return, inflation, etc. If I hadn't seen these posts (and had similar discussions here over the years) and done the trials with spreadsheets I doubt I'd have really come to grips with it.
These exercises also reinforced (to me) the importance of being flexible as we execute the plan. Given the unknowns and unknowables, taking X% of the portfolio year-end-value (or some smoothed variant of the same thing) makes much more sense (to me) than setting a percent at year one and simply adjusting for inflation. That portfolio will be doing something, ignoring it isn't a good policy.

 

[Independent]

Well, granted, if your entire portfolio consists of just 3 Vanguard ETFs and Social Security, then it may not be that involved.

But if you have a more complex portfolio (preferred stocks, individual stocks, etc.), some of which might be Qualified Dividends, maybe some MLPs, some individual REITs that more often than not spin off some Return Of Capital, a large IRA that you might want to either start drawing down now to reduce RMD size and tax bracket, or postpone and take SS now....all of that can be impacted by what you hold in your tax-deferred/ROTH accounts, and what you hold in your taxable account.

In my accumulation years, I put all of my large cashflow producing holdings into my tax-deferred accounts (business development companies, preferred stocks, high yielding individual stocks) to minimize my taxable income, but when I need to live off of the dividends, it will pay for me to spend just a little bit of time tweaking what I hold in which account to maximize the Qualified Dividends that are in my taxable account to live off of, and keeping the regular dividends in the tax-deferred accounts. But that is just one part of the overall tax puzzle.

Then add in the factor of health subsidies and 400% FPL, and you really have a spreadsheet fest of 'what ifs'.

I thought this might be the case. We have a wide range of people on this board. Some are ER due to unusually high net worth, some are ER due to unusually careful LBYM. We have a wide range of investing complexity, from KISS to "some of everything".

I find myself agreeing with samclem today, but I also want to allow for the possibility that I've just missed something.

 

[rayvt]

Good post. Given the availability of time after ER to indulge in these fun "what if" pursuits I think there is a great tendency to measure and try to model things things with the precision of a micrometer when life events are in fact delivered with all the delicacy of an ax.

I would think that if you do want to spent time and effort creating & working on/with spreadsheets, that fiddling with tax-related issues just doesn't have that much value. Seems to me that the effort would be better spent by examining ways to increase your investment returns and/or lowering risk -- such as things that folks like Mel Faber, Gary Antonacci, etc. write about.

I don't see the "save thousands on taxes", either. At 15% rate, $1000 is a $6700 difference in AGI. So for "many thousands", we're talking about differences in the range of $20K-$25K and up. Do people really have the ability to shave that much off of their AGI by twiddling with a spreadsheet?

 

[pb4uski]

....I don't see the "save thousands on taxes", either. At 15% rate, $1000 is a $6700 difference in AGI. So for "many thousands", we're talking about differences in the range of $20K-$25K and up. Do people really have the ability to shave that much off of their AGI by twiddling with a spreadsheet?

If you are planning on managing your income for Obamacare health insurance subsidies beginning in 2014 then the implications of a little additional AGI are huge.

As an example, if I mess up and have $6,700 too much, then while the tax bite is only $1,000 the lost subsidy is $9,700 so that extra $6,700 in income costs me $10,700, 160% of the additional $6,700 in income!!

For me, if I were to increase my income with all 0% capital gains from 400% FPL to the top of the 15% tax bracket (~$40k of gains), while my additional federal income tax would be zero but between state income tax, lost state property tax relief and lost Obamacare subsidies the incremental economic cost is over $14k (~36% of the incremental income).

 

[NanoSour]

If you are planning on managing your income for Obamacare health insurance subsidies beginning in 2014 then the implications of a little additional AGI are huge.

As an example, if I mess up and have $6,700 too much, then while the tax bite is only $1,000 the lost subsidy is $9,700 so that extra $6,700 in income costs me $10,700, 160% of the additional $6,700 in income!!
.

That's encouraging!

 

[dtbach]

ACA is going to make millions for tax advisers it appears. One more thing to game. I didn't realize that the cut off points would be so course as to gain/lose almost $10k!

 

[Midpack]

ACA is going to make millions for tax advisers it appears. One more thing to game. I didn't realize that the cut off points would be so course as to gain/lose almost $10k!

Enough to make you rethink post #44?

Admittedly, it's pointless to go to too far with so many unknowns, but IMO there is value in a sensible happy medium between nth degree and why even bother. Seems prudent to calculate what we can using knowns and reasonable assumptions, and updating when new knowns or assumptions come along. It doesn't take that much effort...

 

[dtbach]

Enough to make you rethink post #44?

Admittedly, it's pointless to go to too far with so many unknowns, but IMO there is value in a sensible happy medium between nth degree and why even bother. Seems prudent to calculate what we can using knowns and reasonable assumptions, and updating when new knowns or assumptions come along. It doesn't take that much effort...

And being a hypocrite I have stuff on spread sheets, check in on VG and Fido regularly, listen to webcasts, etc. So yeah I spend time pouring over stuff. It's just I find that there is only so much you can juggle around (at least in my situation) so after all my research/spread sheet study/etc it has turned out to be a few basic moves and then to use the 4% rule (although again being a hypocrite, I adjust that also as needed)

 

[rayvt]

When I retired, I set up a spreadsheet to compute our annual withdrawals. A bit more complicated than most, since I also adjust the increase by using the Guyton-Klinger rules as well as inflation. Each January I refigure for the next year and get the monthly amount.

But we don't need the same amount each month -- more when we go on a cruise, less when we stay home.
So I have another worksheet where I track the actual withdrawals each month and the running average YTD and the necessary amount for the remaining months.

 

[bo_knows]

When I retired, I set up a spreadsheet to compute our annual withdrawals. A bit more complicated than most, since I also adjust the increase by using the Guyton-Klinger rules as well as inflation. Each January I refigure for the next year and get the monthly amount.

But we don't need the same amount each month -- more when we go on a cruise, less when we stay home.
So I have another worksheet where I track the actual withdrawals each month and the running average YTD and the necessary amount for the remaining months.

I've actually always wondered this about folks who use more complicated variable-spending rules. So, you basically use that Guyton-Klinger number as an upper-limit to spending, and leave the rest that you don't use in your portfolio? It's not like you see a G-K spending level of $70k, up from $65k the year before and say to yourself "Welp honey, looks like we need to spend some more money!", right?

 

[pb4uski]

I guess that I'm not sophisticated enough. I just have a monthly transfer from cash to checking that covers most of our regular living expenses and then transfer additional funds when needed to cover special expenses (vacation, major repair, new car, etc.).

The transfers for a year is what I withdrew and I periodically reassess whether I "have enough" as if I was retiring anew today. If my "have enough" calculations start to creep into a zone of discomfort I would scale back on discretionary expenses, if it goes the other way (as it has since I retired 20 months ago) I allow us an occasional splurge.

 

[Major Tom]

I'm in the same boat as pb4uski. I transfer a fixed amount monthly from my cash reserve savings account to checking. I'm only into my 3rd year of doing this and haven't even increased the amount for inflation yet. Somehow, I have managed to accrue a nice balance in checking over the last few years which will cover many extra expenses such as dental work that is not covered on my plan, vacation etc.

The only big extra expenditure I foresee in a few years may be the purchase of a nice, used RV. For that, I'll just take the cash out of the main stash.

 

[Chuckanut]

Enough to make you rethink post #44?

Admittedly, it's pointless to go to too far with so many unknowns, but IMO there is value in a sensible happy medium between nth degree and why even bother. Seems prudent to calculate what we can using knowns and reasonable assumptions, and updating when new knowns or assumptions come along. It doesn't take that much effort...

+1

Like my old Grand Pappy used to say "Never let the perfect become the enemy of the good." Or to rephrase it for this discussion "Don't let analyzing to the nth degree become the enemy of making some prudent calculations."

 

[Major Tom]

+1
Or to rephrase it for this discussion "Don't let analyzing to the nth degree become the enemy of making some prudent calculations."

+1 from me as well.

However, for those who are of a more analytical mindset, I do enjoy reading the discussions that result. One of the great things about this forum is that the different approaches are rarely promoted with religious fervor; there is an air of genuine scientific interest.

At the very least, if any of us are looking down on others for their approach, we have the decency to keep it to ourselves

 

[FIREmenow]

+1

Like my old Grand Pappy used to say "Never let the perfect become the enemy of the good." Or to rephrase it for this discussion "Don't let analyzing to the nth degree become the enemy of making some prudent calculations."

I had an old boss that used to say, "A good plan today is better than a great plan tomorrow."

 

[rayvt]

I guess that I'm not sophisticated enough. I just have a monthly transfer from cash to checking that covers most of our regular living expenses and then transfer additional funds when needed to cover special expenses (vacation, major repair, new car, etc.)./QUOTE]

I started out doing it that way when I first retired (7 years ago). It's probably a reasonable way to start off while you are getting used to the whole "no paycheck" thing.

But it soon becomes apparent that it's nothing but mental accounting. "Move money to a savings account that I pretend is part of my portfolio, then once a month move money from the savings account to the checking account. And when/if we don't spend it all by the end of the month, move the excess back to the savings account and make a note that we can take it back when/if a month comes when we need a little extra. Oh, and also mentally note than some of the money in the savings account is earmarked for new car/roof/furnace, etc."

Or do you move the excess to a different savings account so you don't get the money mixed up? And do you also have another savings account for the money you'll need for infrequent special expenses? Suddenly you've got half a dozen savings/checking accounts, each one earmarked for a different category of spending.

And then you realize the absurdity of it all. All you are doing is moving money from one savings account to another one. What's the point of that?

It seemed simpler to me to approach it from the opposite direction. Instead of funneling out a specific amount money and then directing it to different usages, keep running track of the expenditures and generally keep it down to the amount that your SWR calculation says.

Normal method, similar to what you do when your income is a paycheck, is get a constant monthly income and parcel it out to expenses.
My method is to pay expenses and vary the monthly income, under the constraint that the annual income cannot be more the $X.
"Income" = withdrawal from portfolio.

It's just a lot easier. If in Feb you have a large expense (new car, luxury cruise, etc.) you pay it, and cover it by reducing the allowed spending for Mar-Dec. If you didn't spend much from Jan to Jul, you can splurge in Aug (or small splurge in Aug-Dec) because the money you didn't take is still there.
The target is to end in December with the average monthly withdrawal for the last 12 months to be equal to the amount that your SWR says.

It's easy to set up in a spreadsheet. You only have to keep track of one thing -- the money you transfer to your checking account and then spend. Computing the running average since BOY and the to-go-to-EOY average is simple.

 

[teejayevans]

My method is to pay expenses and vary the monthly income, under the constraint that the annual income cannot be more the $X.
"Income" = withdrawal from portfolio.

It's just a lot easier. If in Feb you have a large expense (new car, luxury cruise, etc.) you pay it, and cover it by reducing the allowed spending for Mar-Dec. If you didn't spend much from Jan to Jul, you can splurge in Aug (or small splurge in Aug-Dec) because the money you didn't take is still there.
The target is to end in December with the average monthly withdrawal for the last 12 months to be equal to the amount that your SWR says.

It's easy to set up in a spreadsheet. You only have to keep track of one thing -- the money you transfer to your checking account and then spend. Computing the running average since BOY and the to-go-to-EOY average is simple.

My method is even easier, I use SWR as a guideline, if I go over SWR because of a big ticket item, no worries. I just check my withdrawals at the end of the year to see how much I spent. No spreadsheet of expenses, no trying to cut expenses for the rest of the year...hey this is retirement, it's suppose to be fun , tracking my expenses with spreadsheets and forcing myself to stay under some arbitrary limit, that's no fun (Note if the economy tanks again, I would just postpone any big ticket item.)

 

[Huston55]

When I retired, I set up a spreadsheet to compute our annual withdrawals. A bit more complicated than most, since I also adjust the increase by using the Guyton-Klinger rules as well as inflation. Each January I refigure for the next year and get the monthly amount.

But we don't need the same amount each month -- more when we go on a cruise, less when we stay home.
So I have another worksheet where I track the actual withdrawals each month and the running average YTD and the necessary amount for the remaining months.

That is also our plan, to follow a VWR approach using Guyton-Klinger rules to guide withdrawals, as discussed in this thread. http://www.early-retirement.org/for...tor-ive-seen-to-date-68770-4.html#post1373036

Using a 'floor & ceiling or 'guardrails' approach like the 'Guyton-Klinger' enables higher beginning WDR. It also has the possibility of a lower WDR but, provides the structure for making adjustments as time and circumstances occur, with some knowledge of where it all ends up. I like this combo of discipline and flexibility.

For example, with our AA, we forecast surviving a 40 yr period and could begin with a 5.3% 'real' WDR (we plan to start with 5.0%), which has a worst case possibility of reducing to a 'real' 3.7% WDR over an eight year period but, also has the possibility of increasing to a WDR >5.3%.

3.7/5.0=76% So, 24% is a large % decrease in annual income, if the worst case comes to pass. However, in our situation, we have some pension income that comes on-line about two years into FIRE, which would temper that reduction. So, G-K seems to be a good fit for our situation. Benefits of the approach include:

1. Higher starting WDR earlier in RE when it's most useful.
2. An approach that is flexible to accommodate bad sh!t happening
3. An approach that provides a disciplined structure, which I think we need.

Rayvt-I'd be interested in hearing how G-K has worked for you thus far.

 

[Danmar]

Just read this thread from start to finish. Interesting. Our financial affairs seem pretty simple compared to some here. We just spend our pensions and dividends. Keep a fairly large cash balance to absorb lumpy expenses like new cars or expensive trips. Vitually all portfolio is in taxable accounts. Tax planning pretty simple as we pay at the max marg rates and always will. If I was in the same position as some of you, I think I would spend a lot of time analyzing taxes as well.
I am probably leaving some money on the table as I manage all our financial affaires myself but I like doing it and this keeps if simple and understandable.