Pfau: https://www.onefpa.org/journal/Pages/OCT15-Making-Sense-Out-of-Variable-Spendin

[walkinwood]

Pfau examines various variable spending methods and compares them using consistent assumptions.

https://www.onefpa.org/journal/Page...ariable-Spending-Strategies-for-Retirees.aspx

It is an interesting read and I recommend it to people just starting to think about ER and the method(s) they will use to withdraw money in retirement.

However, in my 8th year of ER, I'm getting tired of reading these papers.

I am beginning to feel confident that our 4% of portfolio value on Jan 1 method can work with a bit of flexibility on our part. This method can be volatile, so a fair amount of head-room in your initial annual budget will make it easier to stomach. Our portfolio is 60 equities (large value/growth, small, international & REIT) and 40% bonds (TIPS and short/intermediate treasuries & corporate bond funds)

We also put aside a bit of money that we do not use to calculate our withdrawals. It acts as a buffer against volatility of withdrawals and is there in case we have some kind of emergency requiring gobs of money.

 

[2017ish]

Agree that it is a good overview and I've already passed along to a couple of people.

We're planning to take your approach of constant percentage withdrawal, which skips a lot of the issues--as long as, like you say, we have a [large?] bit of flexibility.

 

[W2R]

However, in my 8th year of ER, I'm getting tired of reading these papers.

I'm feeling the same way after 6 years of retirement. But I still bookmark the papers just in case.

The market has been so doggone good for the past 6 years since I retired, that I got a huge shove in the right direction early in retirement. As my portfolio spiraled upwards, with spending remaining about the same, my WR dropped and dropped. This year when it got closer to 1.5% than 2% of the January 1st value, I bought a house (which blew that to smithereens) but it will still be under 2% next year. Still, I find that in good times like these I run out of spending ideas even though I planned on variable withdrawal.

When the market tanks, which we KNOW it will , we will all be thinking of cutting way back, I suspect. I'll be much more worried about WR and may dredge up these papers at that time, and review them.

 

[walkinwood]

I agree, W2R. I still read the papers and we've been living very happily on less than the allotted 4%, so we *should* be fine when the market dives as it surely will at some time.

 

[ejman]

Between taking early SS (I know, I Know) and taking all distributions (dividends and CG) in cash from the taxable accounts (about 41% of liquid NW) we find that our monthly expenses are pretty much covered so I've pretty much stopped reading the "how to take distributions in retirement" treatises. But as mentioned by W2R I'm sure that if the market/economy dives the topic will again become much more interesting to me.

 

[Lsbcal]

Pfau examines various variable spending methods and compares them using consistent assumptions.

I'll try to look at this but there are just too many ways that these things have changed. Why doesn't he just find the ultimate and publish that? Or he could say what he favors and always test a new idea against that.

We also put aside a bit of money that we do not use to calculate our withdrawals. It acts as a buffer against volatility of withdrawals and is there in case we have some kind of emergency requiring gobs of money.

This gives me an idea. I'll start tracking the cumulative amount we've missed on the year's spending target. Should help emotionally in a declining portfolio and bad market year.

This year I decided VPW was a really good way to do things. Forced me to confront my inner frugal demon. And it is all there in a spreadsheet versus a paper with all kinds of academic ideas that one must hope have been examined rigorously.

This year VPW says (with a conservative end of age of 110) that we can spend 4.5% of the portfolio. Might only spend about 3.4% even with nice vacations and a big home project. So that extra 1.1% would go into the "could have but didn't spend" bucket.

 

[walkinwood]

This gives me an idea. I'll start tracking the cumulative amount we've missed on the year's spending target. Should help emotionally in a declining portfolio and bad market year.

I don't understand your statement.Would you please elaborate?

 

[Mo Money]

Walkinwood, I think Lbscal intends to add up what budgeted spending money was not spent each year, so that that buffer amount makes Lbscal feel better when financial times are tough. But Lbscal can confirm or correct my interpretation.

I do a version of the same thing. I keep track of what budgeted money was not spent each month and set it aside (on paper) so that I can tap into it later in the same calendar year, should I need it to stay on budget for that year. If there's an excess at the end of each year, I'll know how much I did not spend of my budgeted amount. Any excess will be used to help finance the next calendar year's budget, along with CG and dividends.

I am planning to withdraw a bit less than 3% annually of my balance on ~January 1 of each year. I am aiming to use that 3% target for about 6 years till I turn 60. Then I'll revisit my percentage and maybe bump it up to 3.5%. Or not.

If I read every one of Pfau's articles, I would have no time for retirement!

 

[nun]

What jumps out at me is the inappropriateness of the "Fixed Percentage" withdrawal (the third strategy tested), even though it removes sequence of return risk by automatically reducing spending when the value of the portfolio falls. However, the volatility of the income stream really makes it impractical in the real world.

 

[big-papa]

What jumps out at me is the inappropriateness of the "Fixed Percentage" withdrawal (the third strategy tested), even though it removes sequence of return risk by automatically reducing spending when the value of the portfolio falls. However, the volatility of the income stream really makes it impractical in the real world.

Burn's and others modified that method to minimize the drop year on year to some fixed percentage. The one he published was to withdraw either the larger of 6% of the portfolio or 90% of the previous year's withdrawal, whichever is larger. Nothing magic about those two numbers - you can choose anything that makes you happy or can be justified by at least some backtesting.

 

[Lsbcal]

Walkinwood, I think Lbscal intends to add up what budgeted spending money was not spent each year, so that that buffer amount makes Lbscal feel better when financial times are tough. But Lbscal can confirm or correct my interpretation.
...

Yes, that is what I was thinking. I personally have a lot of fears about sudden major declines (like 2008) and loss of wealth. I've had to constantly remind myself that it is not a virtue to hoard money.

VPW seemed to have a clear methodology on the spending side and luckily the numbers it comes up with are well within our spending needs. They are actually somewhat over our spending needs, so that spending shortfall should be tracked.

 

[CaliforniaMan]

What jumps out at me is the inappropriateness of the "Fixed Percentage" withdrawal (the third strategy tested), even though it removes sequence of return risk by automatically reducing spending when the value of the portfolio falls. However, the volatility of the income stream really makes it impractical in the real world.

Some of the income volatility I am sure comes from the over 12% initial withdrawal rate. Maybe I am missing something in my quick read, but that just seems nuts. My own plan (I am only 18 months in) is a fixed withdrawal of 3.5%. Even though I am 70/30 I calculate that a 30% stock market fall will decrease my spending by only 15%. Taking into account my SS income and reduced taxes. With no mortgage and small fixed expenses that doesn't seem extreme to me.

And if that happens, I really do have 30% less than before, so if I were just beginning my withdrawals that is what I would take, so why take more now. I know that based on past history this is very conservative, but we don't know the future. The past is a good yardstick by which to make a guess, but obviously we don't know the future, so hence the being conservative reasoning.

I think it depends on your situation, whether you have another annuity or pension or SS stream, your withdrawal percentage, and how large your fixed expenses are.

For me it just seems a fixed percentage makes sense. And it is simple. I like simple.

 

[nun]

I think it depends on your situation, whether you have another annuity or pension or SS stream, your withdrawal percentage, and how large your fixed expenses are. For me it just seems a fixed percentage makes sense. And it is simple. I like simple.

Yes these withdrawal strategies are just the starting point and need to be modified for personal circumstances. I imagine lots of people want to leave some inheritance, I certainly do.

Right now I'm just spending cash until my pension starts in 12 months time. So my withdrawal from my cash account is a 1/N strategy per month with N as 18 rather than 12 to give me a safety margin. When my spending is covered by a pension and rent my withdrawal strategy from mutual funds will be as simple as taking some dividends or interest from TIAA-Traditional if I need extra cash. I also like to keep it simple.

 

[hnzw_rui]

Yes these withdrawal strategies are just the starting point and need to be modified for personal circumstances. I imagine lots of people want to leave some inheritance, I certainly do.

Right now I'm just spending cash until my pension starts in 12 months time. So my withdrawal from my cash account is a 1/N strategy per month with N as 18 rather than 12 to give me a safety margin. Because my spending will soon be covered by a pension and rent my withdrawal strategy from mutual funds is limited to taking some dividends or interest from TIAA-Traditional if I need extra cash. I also like to keep it simple.

+1. I'll probably do the same. I guess the closest scenario for me will be annuitize floor + invest for discretionary. Current plan is to use pension as spending floor (and maybe use some 457 funds to buy higher pension). Then every year before RMD kicks in, take 12-20K distribution from 457 (not to exceed 25% bracket) to either spend or convert to Roth. Not planning on being aggressive with the Roth conversions. If returns are so good that RMDs put me in the 33 or 39.6% bracket, I'll just count myself lucky.

 

[audreyh1]

What jumps out at me is the inappropriateness of the "Fixed Percentage" withdrawal (the third strategy tested), even though it removes sequence of return risk by automatically reducing spending when the value of the portfolio falls. However, the volatility of the income stream really makes it impractical in the real world.

Not impractical at all. I manage just fine.

 

[hnzw_rui]

Not impractical at all. I manage just fine.

Question though, do you have a bigger bond/fixed income allocation to minimize portfolio volatility for more predictable withdrawals?

 

[audreyh1]

Some of the income volatility I am sure comes from the over 12% initial withdrawal rate. Maybe I am missing something in my quick read, but that just seems nuts. My own plan (I am only 18 months in) is a fixed withdrawal of 3.5%. Even though I am 70/30 I calculate that a 30% stock market fall will decrease my spending by only 15%. Taking into account my SS income and reduced taxes. With no mortgage and small fixed expenses that doesn't seem extreme to me.

For me it just seems a fixed percentage makes sense. And it is simple. I like simple.

Me too.

Not only that, but if most of your investments are in taxable accounts, taxes tend to be higher in years when the portfolio grows. If it shrinks, taxes often drop. So after tax spending is not hit as hard in the down years. This lessens the volatility somewhat. In the real world I've seen this effect first hand tracking my portfolio for 15 years now.

 

[audreyh1]

Question though, do you have a bigger bond/fixed income allocation to minimize portfolio volatility for more predictable withdrawals?

Well - I have a target allocation - ~53/47. I picked it for lower volatility overall plus good long-term portfolio survival. I can't say that I picked it for more predictable withdrawals, although it obviously reduces the volatility. Allocations in that ballpark are recommended for other withdrawal methods too so I don't see the 47% fixed income as a "bigger allocation".

OK - I'll disclose my big secret. I practice a form of "spending smoothing". If my portfolio has a big increase one year, say 10%, my withdrawal increases by 10% - that's like a 10% raise.

But I don't automatically increase my spending by 10%, but let it increase more gradually over the years. I let the excess build to help fund future years when I have to take a "pay cut". This is just being prudent, IMO. Lean years, fat years, etc.

 

[Lsbcal]

I talk in percentages and so do a lot of people here. But just want to mention that one person's 3% is not the same as another's i.e. a $10M portfolio base is a lot different then a $1M base. So cutting back spending in bad years will be much more painful for some then others.

 

[Lsbcal]

Well - I have a target allocation - ~53/47. I picked it for lower volatility overall plus good long-term portfolio survival. I can't say that I picked it for more predictable withdrawals, although it obviously reduces the volatility. Allocations in that ballpark are recommended for other withdrawal methods too so I don't see the 47% fixed income as a "bigger allocation".
...

We have similar AA's.

 

[nun]

+1. I'll probably do the same. I guess the closest scenario for me will be annuitize floor + invest for discretionary. Current plan is to use pension as spending floor (and maybe use some 457 funds to buy higher pension). Then every year before RMD kicks in, take 12-20K distribution from 457 (not to exceed 25% bracket) to either spend or convert to Roth. Not planning on being aggressive with the Roth conversions. If returns are so good that RMDs put me in the 33 or 39.6% bracket, I'll just count myself lucky.

Yes my plan is to buy into my state's pension with 403b and 457 funds and that ,along with rental income, provides a generous income floor for my frugal lifestyle. I'll do IRA to ROTH conversions from age 55 to 70 up to the top of the 15% tax bracket. My plan is to use the ROTH as the core of my estate to leave to my nieces in the UK as its tax free in both the US and the UK. I'll take dividends and TIAA-Traditional interest from taxable and 401a accounts if necessary, always trying to stay inside the 15% tax bracket to avoid tax on the dividends.

 

[walkinwood]

Not impractical at all. I manage just fine.

+1.

Having headroom in your spending is essential.

If all your expenses are non-discretionary, a drop in your portfolio will send you back to work.

 

[Lsbcal]

+1.

Having headroom in your spending is essential.

If all your expenses are non-discretionary, a drop in your portfolio will send you back to work.

Good point about "non-discretionary". Actually, come to think of it, that was kind of what I was getting at above.

We should probably all be mentioning the % discretionary in our numbers when sharing here. But then again, that might get into "envy" territory. And some of us don't have expensive tastes even if we could afford it. For example, I never buy wine much above $9 a bottle.

 

[audreyh1]

+1.

Having headroom in your spending is essential.

If all your expenses are non-discretionary, a drop in your portfolio will send you back to work.

And retiring early is not a good idea if you have no slack.

Someone considering fixed percent might want to be able to handle a 25% drop in withdrawn income any given year or have built up a buffer to help out in years where the withdrawal drops.

 

[hnzw_rui]

OK - I'll disclose my big secret. I practice a form of "spending smoothing". If my portfolio has a big increase one year, say 10%, my withdrawal increases by 10% - that's like a 10% raise.

But I don't automatically increase my spending by 10%, but let it increase more gradually over the years. I let the excess build to help fund future years when I have to take a "pay cut". This is just being prudent, IMO. Lean years, fat years, etc.

That seems a great idea. Curious, where do you keep the extra withdrawals? Just cash or cash equivalent?

And retiring early is not a good idea if you have no slack.

Someone considering fixed percent might want to be able to handle a 25% drop in withdrawn income any given year or have built up a buffer to help out in years where the withdrawal drops.

I think for someone planning to ER, having a nest egg of 30-35x of essential expenditures might be prudent. Granted, that doesn't take SS or pensions into account.