Combo Withdrawal Strategy

[lark_L]

Been thinking a lot about withdrawal strategies. Of course, the constant % of portfolio or constant % inflation adjusted are thoroughly discussed and documented. I've been reading about the 1/n method Withdrawal Methods - Bogleheads

The issue for me with the % methods is that after the first 10 years or so, you're still withdrawing the same amount, when you could likely increase it safely. We have no kids, so we're not worried about leaving anything.

The 1/n method looks more appealing - as long as you don't select too low an "n" - which would be bad. But, the withdrawal amount is quite low in the beginning. Arguably, so low that you'd either not be able to retire or you couldn't conform to it, unless you were quite well off (and then you wouldn't care either way).

So, now I'm thinking that I combine the two. Do the calcs for both each year, then take the maximum of either one. What this does is the following - you take the % amount out until the 1/n amount is larger, then you take that out for the rest of your life.

Maybe it's too complicated and I guess you could just start out at 3-4% and just increase it later after you get through (hopefully) the first - very important - years.

 

[rayvt]

As I read it, the 1/n method assumes its conclusion.
It actually said so: "...N number (i.e. how long you will live) can't be forecast ..."

Well, DUH!!! I we knew how long we were going to live, all this would be really easy.

Take a look at the Guyton/Klinger withdrawal strategy. That makes sense, is easy to comprehend, and doesn't have wild swings in the withdrawn dollar amounts.

 

[braumeister]

Maybe it's too complicated and I guess you could just start out at 3-4% and just increase it later after you get through (hopefully) the first - very important - years.

Very true; the later years are not nearly as important, and can be ignored for now.

No matter how old I get, I'm sure refrigerators will still come in large, sturdy cardboard boxes, and if I can find a nice quiet alley somewhere close to a soup kitchen, I don't expect to need much in the way of assets at all.

See? It doesn't have to be complicated at all!

 

[nun]

I don't really get all these withdrawal strategies. My strategy is to take enough money from my retirement savings to fund my expenses without eating into principal. Therefore, I'll reduce spending and use cash in lean years and save or reinvest any surplus in good years.

It's the "Mcawber" approach and begins with having an investment to expense ratio large enough so that you are not up against the usual SWR.
However, there is some thought required from the tax stand point so that RMDs don't produce big tax bills down the road. So in good years I might withdraw more than I need to live on just to fill up a tax bracket and that put the surplus into after tax investments.

 

[Nords]

Maybe it's too complicated and I guess you could just start out at 3-4% and just increase it later after you get through (hopefully) the first - very important - years.

After several hundred thousand posts on the subject, this is about the only consensus...

 

[meierlde]

Note that an RMD is essentially a 1/n method with N being your remaining life expectancy. If you live to 100 the IRA/401k is empty. So to start with some part of withdrawals will be 1/n (at least to the extent of paying taxes and moving to taxable investments). If one has the time, I suggest downloading the IRS life expectancy tables for RMD's and building a spread sheet to see what they would look like. In particular then one can play with what to do during the period between 59.5 and 70.5 where withdrawals are optional.

 

[nun]

Note that an RMD is essentially a 1/n method with N being your remaining life expectancy. If you live to 100 the IRA/401k is empty. So to start with some part of withdrawals will be 1/n (at least to the extent of paying taxes and moving to taxable investments). If one has the time, I suggest downloading the IRS life expectancy tables for RMD's and building a spread sheet to see what they would look like. In particular then one can play with what to do during the period between 59.5 and 70.5 where withdrawals are optional.

Right now I'm thinking more of 50 to 59.5. I can live off rent and after tax money until I get to 59.5. I'm still thinking about whether to simply roll over amounts up to the 15% tax level from my IRA to ROTH or to 72t the same amount each year. The 72t would mean I'd be locked in, but it would give me more flexibility from a cash flow stand point.

Any opinions?

 

[braumeister]

I suggest downloading the IRS life expectancy tables for RMD's and building a spread sheet to see what they would look like. In particular then one can play with what to do during the period between 59.5 and 70.5 where withdrawals are optional.

Excellent advice.
I did this not long ago, and was most unpleasantly surprised. Didn't realize the RMDs would be quite that large. Interesting project trying to minimize that.

 

[Independent]

Maybe it's too complicated and I guess you could just start out at 3-4% and just increase it later after you get through (hopefully) the first - very important - years.

You didn't give your age. I think that if you're retiring a 50 you do something different than if you're retiring at 70.

The 1/n method essentially says that you an eat into principal. But, it you're young, the amount you can get each year is trivial. You should just assume you'll live forever and take (a conservative version) of your expecting earnings, which works out to the 3-4%.

When you get older, you can start thinking about how much risk you're willing to take, and a calculator like FireCalc can be run with 20 or 10 year horizons. They "allow" you to chew up principal in the bad scenarios.

------

Now, if you really want to get out the sharp pencil, use 1/pva where pva = present value of an annuity that runs as long as you think you might live. If you calculate pva with an assumed interest rate of 0%, you get 1/n. For any higher rate, pva<n and you get to spend more money. If your earnings happen to exactly match the assumed interest rate, you'll get a perfectly level income over your life.

This is the math in an SPIA but the insurance company gets to use average life spans instead of maximum life spans. It's also the basis of the withdrawals in a pay-out variable annuity (a product with a lot of theoretical attraction, but with too much baggage for most of us).

 

[Midpack]

I don't really get all these withdrawal strategies. My strategy is to take enough money from my retirement savings to fund my expenses without eating into principal. Therefore, I'll reduce spending and use cash in lean years and save or reinvest any surplus in good years.

On behalf of your beneficiaries, thanks!

For those of us with no kids who'd rather die broke (and know that's just as impossible to predict) AND who do not want big fluctuations in annual income, we'll have to judiciously 'eat into principal' at some point - consequences and all.

We intend to start at about 2.5% WR and course correct as the years roll by. Life has had it's uncertainties so far, and I suspect that will continue despite my retirement, no matter how inconvenient that might be.

Good luck to all of us, very few of us will get it exactly right.

 

[rescueme]

Maybe it's too complicated and I guess you could just start out at 3-4% and just increase it later after you get through (hopefully) the first - very important - years.

That may be one way, but not the only option especially if you ER and have many years to go until all income sources (such as pensions, SS, etc.) come "on-line".

In our case, we have/will have many years before all those sources are available. While I retired four years ago (at age 59), I won't draw SS till age 70 - primarily for the benefit of my DW.

DW (same age as me) is still wo*king but may retire any day. We certainly fortunate that we don't need the income but she remains employed due to not being emotionally ready to retire. She will have two small defined benefits (e.g. pensions) in two years, when she turns 65. She plans to take SS at her FRA age of 66.

That means that our "guaranteed income sources" consisting of future pensions, each of our SS, my 50% claim of her SS at her FRA, and my maximum SS at age 70 will not occur at retirement, but various dates between my retirement age of 59 till my SS age of 70. That is eleven years of withdrawals from our joint retirement portfolio in excess of that "magic 4%".

We're in trouble, right? No, not really. Using a tool such as FIDO's Retirement Income Planner (which shows year by year income, taxes, RMD's, portfolio value, and most importantly rate of withdrawal) shows that at age 70, our WD rate drops back down to 2.5%. It doesn’t exceed 4% (slightly) till our mid-80's, assuming we live that long.

If I would have held to a 4% withdrawal rate at my retirement at age 59 (in addition to our SPIA income, since that date), two things would have happened. We would have had to reduce our lifestyle upon my retirement. Secondly, even if we would have adhered to that withdrawal rate, we would have had "too much" left on the table in the end, as each income source started.

This is just to cite a real-life example of where a "simple" 4% WD rate may not work, especially for those that retire before all their retirement income starts on the day of retirement.

 

[devans0]

Our plan is to spend down the IRA and convert to non-IRA investments with the excess beyond our needs. This will keep our tax bite to a lower tier, and reduce RMD, assuming tax tiers still exist
Second phase will be to spend the regular investment income money.
Third phase will be Roth, assuming that we get there. Tax wise, this is the efficient path for us.
We do plan on leaving a bunch on the table. A gift and endowment to the future generations.... or not, if the politicians get their way. We will have plenty for ourselves, so it won't mean a sacrifice.

 

[meierlde]

Excellent advice.
I did this not long ago, and was most unpleasantly surprised. Didn't realize the RMDs would be quite that large. Interesting project trying to minimize that.

But yet in another sense its a good problem to have, it means you have reasonable resources.
BTW if one wants to make any chariatable gifts at death, the 401k/ira is an ideal source, as that eliminates the taxes on the amount given. (Even assuming the estate tax limit stays where it is).

 

[lark_L]

Thanks for all the replies, lots to mull over now. This post really hits home, because this is EXACTLY where I'll be. (Need to exceed 4% until other income kicks in)

We're in trouble, right? No, not really. Using a tool such as FIDO's Retirement Income Planner (which shows year by year income, taxes, RMD's, portfolio value, and most importantly rate of withdrawal) shows that at age 70, our WD rate drops back down to 2.5%. It doesn’t exceed 4% (slightly) till our mid-80's, assuming we live that long.

I'm getting the IRS RMD tables now - back to the spreadsheets!

 

[rmark]

You may want to check out an article on Henry Hebelers website

http://www.analyzenow.com/Articles/Planning%20Pre%20Retirrement/Planning%20Pre%20Retirement%20Articles/Financial%20Dynamics%20Improve%20Planning%206-24-04.pdf

he uses a combination of previous portfolio balance plus inflation and an annual division of portfolio - in up years you only get the inflation increase, in down years the annual division, smoothing and stretching withdrawals. And conveniently at 60 (from IRS publication 590), an individual has a 25 year life expectancy, for an initial 4% withdrawal.

 

[chinaco]

...

course correct as the years roll by. Life has had it's uncertainties

.../QUOTE]

That is it in a nutshell....