$300,000 in Three Buckets...to last 30 years from age 66 to

[hudson]

$300,000 in Three Buckets...to last 30+ years from age 66 to 96


Bucket 1 = $124,274, age 66 to 76, $1200/month, assume 3% return, use all
Bucket 2 = $92,888 grows to $167,012 assuming 6% growth, age 76 to 86, $1613/month, use all
Bucket 3 = 82,838, grows to $320,558 assuming 7% growth, renew buckets

Bucket 1 Looking for a 3% return
33.33% Vanguard Prime
33.33% Vanguard Intermediate Term Treasuries
33.33% Vanguard Inflation Protect Sec. Inv

Bucket 2 Assuming a 6% return
33.33% Vanguard Intermediate Term Treasuries
33.33% Vanguard Inflation Protect Sec. Inv
16.67% Vanguard Total Stock Market
16.66% Vanguard Vanguard FTSE All-World ex-US Index

Bucket 3 Assuming a 7% return
35% Vanguard Total Stock Market
35% Vanguard Vanguard FTSE All-World ex-US Index
10% Vanguard Small-Cap Index
10% Vanguard REIT Index Fund
10% Vanguard Long-Term Treasury Fund

How can I improve? I plan to read the other books that CyberBob recommended.


Thank you CyberBob and Ray Lucia (Buckets of Money)!
Note: The asset allocation strategy is from Diehards/Bogle/Ferri/Swedroe/Larrimore/Lindauer/Lebouef/Bernstein.

http://www.diehards.org/forum/viewtopic ... 54df6fed6a
http://www.amazon.com/Buckets-Money-Ret ... 728&sr=1-1

[market timer]

Have you considered inflation?

[hudson]

"Have you considered inflation?"

Somewhat...From Bucket 1 to Bucket two is an increase for inflation from $1200 per month to $1613 per month. Inflation protected bonds are included as well as stock mutual funds.

[market timer]

What happens to bucket 3 when you're 85? Still invested 90% in equities?

In terms of your returns, they look reasonable in expectation. The second bucket is probably expected to return less than 6%, while the first and third might be expected to return a little more than your targets of 3% and 7%.

[markh]

What am I missing?

isnt 3 percent of 124,000 about 3700 per year? how does it come out to 1200 per month

mark

[nisiprius]

Primary Annuitant -- Birth date: 01/01/1942 Sex: M
Quote Expiration Date: 02/18/2008
Benefit Commencement Date: 03/19/2008
State of Residence: IA
Payments per Year: 12
Total Premium Amount: $300,000.00

Initial Payment Amount for Fixed Single Life Annuity with inflation adjustments: $1,573.68

That is, for a man aged 66, $300,000 in an inflation-adjusted single-premium annuity from Vanguard/AIG will pay $1,573.68 per month initially, increasing with the CPI index thereafter. ($1,415.16 for a woman; $1,225.52 for an opposite-sex couple, joint-and-survivor).

No assumptions about rates of return, no assumptions about inflation, pay 'em $300,000 and they'll pay you $1,573 year-2008 dollars a month for the rest of your life, no mutual-fund-prospectus weasel wording about "might not achieve its goals," it's a contract. If there's a bear market or the Treasury quits issuing TIPS it's their problem, not yours.

That's not a recommendation. But it is something to think about and consider in planning. If you've already thought about it and rejected it, fine.

I don't want to go into a detailed song-and-dance about all of the reasons why you might not want an SPIA. The big difference from a do-it-yourself investment is that with an investment, if you live too long you run out of money but if you die young your heirs get what's left over; with an SPIA, you can't run out of money but there's also no such thing as having anything left over. And even the insurance companies warn you not to put more than half of your assets into an SPIA. And the insurance company might go bust, or the BLS might fudge the CPI to the point of making CPI adjustments useless, etc.

I used Vanguard/AIG for the illustration because they have exact quotations online. I don't sell insurance and I don't work for Vanguard. (I keep tossing these quotations in so often that I worry that people might think that I do).

[cherijoh]

What am I missing?

isnt 3 percent of 124,000 about 3700 per year? how does it come out to 1200 per month

mark

You are assuming spending interest only; the original poster is spending principal.

C

[Rich_in_Tampa]

I think Ray would say:

Bucket one probably shouldn't have intermediate term bonds or even TIPS (safe as they are, they can fluctuate in value when owned through a mutual fund). Nothing longer than short term bonds is the guide there.

Bucket 2 should have nothing more volatile than something along the lines of a 60:40 balanced fund, so I would take the total stock market index and the FTSE EFT out of bucket 2. I like intermediate bonds, TIPs, and Wellesley here, by the way.

Bucket 3 is where your total stock, international idx, REIT fund, etc. belong and remain untouched for, say,14 years. Only one I don't see fitting in there is your LT bond fund - this is a growth bucket.

Yes, you burn through or "self-annuitize" bucket 1, then move bucket 2 to bucket 1 and repeat. In the interim you can do a little light rebalancing from 3 to 2 or 2 to one when returns beat expectations. But largely you leave bucket 3 alone. I am oversimplifying but that's the gist of it.

At the end of say 14 years you are almost entirely in stocks, but the 14 year hold virtually assures that you are well ahead at least historically. The long term growth, trumps any current ups and downs of the market when you are finally ready to reallocate that bucket. Just replenish buckets 1 and 2 using bucket 3 according to the original formulas (present value calculations based on historic returns for each bucket).

The sequence of withdrawing cash first, then bonds, then stocks as a withdrawal strategy (versus conventional pro-rata withdrawals with rebalancing, or selling whatever is high at that time) has support in academic studies - Spitzer et al, Journal of Financial Planning, June 2007.

While I am not a 100% adherent to the Lucia plan, I have something pretty close. Bogel meets Lucia, you might say. It is NOT for everyone, but I rather like it.

[4th&Goal]

I like the buckets strategy and Grangaard and Seguptha too (Though I may have butchered the spelling).

[mikenz]

While I am not a 100% adherent to the Lucia plan, I have something pretty close. Bogel meets Lucia, you might say. It is NOT for everyone, but I rather like it

I'd like to hear more. I'm 15 years from retirement, but have read Lucia and Grangaard books from the library and was intriqued. I especially wonder about the practical aspects - I don't imagine you would literally separate buckets 2 and 3 for example into physically separate funds, but would just account for them in a spreadsheet or something?

[Skiffy]

Which retirement books talk about the "bucket" system for retirement spending?

[CyberBob]

Which retirement books talk about the "bucket" system for retirement spending?

The major bucket-type books would probably be:
Buckets of Money, Lucia
The Grangaard Strategy, Grangaard
The Only Proven Road to Investment Success, Sengupta
Asset Dedication, Huxley

And although not the primary focus of the book, I also like the Liquidity Test section on page 173 of Larry Swedroe's The Only Guide to a Winning Investment Strategy You'll Ever Need

Bob

[JDeeds]

Frank Armstrong proposed a similar strategy, but different in many ways from Lucia. Basically, he proposed that you keep 7-10 years of anticipated expenses in ST Bonds to draw from and the remainder in stocks. In good years, you would replenish your bonds portfolio from stock gains. The portfolios would be managed separately. This is very close to my plan, which I am implementing this year, although I am somewhat more conservative as this would result in a higher level of stocks than I am comfortable with.

http://www.investorsolutions.com/v2cont ... rement.pdf

[Rich_in_Tampa]

I'd like to hear more. I'm 15 years from retirement, but have read Lucia and Grangaard books from the library and was intriqued. I especially wonder about the practical aspects - I don't imagine you would literally separate buckets 2 and 3 for example into physically separate funds, but would just account for them in a spreadsheet or something?

I definitely do divide the buckets into separate funds, because such separation is inherent in the strategy. The Buckets book by Lucia is easy to read (skip his second book called Ready, Set, Retire), IMHO.

The basics are as per my last post: bucket one is cash and near-cash (short term bonds at the most). You start with anywhere from 5 to 7 years worth of annual expenses including inflation corrections. You burn through that including principle.

Bucket 2 is for investments which historically reliable beat cash, but need about 7 years of maturity before that becomes close to a sure thing. That is, a little more reward with the attendant risk neutralized by the 7 year hold. When Bucket 1 runs dry, you replenish it from bucket 2. Balanced funds, bonds, Wellesley, etc go well here.

Bucket 3 is a diversified stock portfolio. Because it sits for 14 or 15 years, it has enough time to reliably allow you to start the whole process over by reallocating all buckets at that time.

I am oversimplifying hopelessly, but the numbers, calcluations and historic returns seem to work for me. There is, in fact, some interim rebalancing but not much.

Note that theoretically after B1 and B2 are gone you are 100% in stocks. In reality, most would keep a couple of years of fixed investments on hand at all times to help you choose your timing a bit more selectively. The 100% stock allocation doesn't last long and is not as scarey as it sounds since that money has been there growing for > 14 years and regardless of the prevailing market at that time it should be plenty big to make it all work quite safely.

What I don't like about it: he pushes annuities a little bit in the book; same with non-traded REITs which I am wary of; the portfolio does migrate inexorably toward stocks, but you can adjust this by deciding how many years you want in B1 - the fewer years, the more in stock; it is a little complicated until you spreadsheet all of it; he does not make it clear that the initial balances for B1 and B2 are actually present value calculations, corrected by inflation (keeps it mysterious but you can easily spreadsheet it); you are looking at a huge chunk of cash when you start out (e.g. 25% of holdings) so you have to remember that melts away each year (you don't rebalance B1, you just drill right through it for 7 years).

My investments are a standard mix of stock index funds led heavily by TSM for B3; a mix of TIP fund, Wellesley and TBM for B2; and short-term fed bonds and MMF for B1.

Sorry to ramble, hope that helps.

[mikenz]

Thanks for that.

What split do you have between taxable and tax-advantaged? Do you hold the stocks in taxable? If so you'd be drawing down the tax-advantaged accounts it seems.

Also, if the stocks in bucket 3 did really well for a few years, would you siphon off some of the gains prematurely, or simply stick with the plan for the full 14 years?

[Rich_in_Tampa]

Thanks for that.

What split do you have between taxable and tax-advantaged? Do you hold the stocks in taxable? If so you'd be drawing down the tax-advantaged accounts it seems.

Also, if the stocks in bucket 3 did really well for a few years, would you siphon off some of the gains prematurely, or simply stick with the plan for the full 14 years?

I have all my post-tax dollars in bucket 1. If I had more than I needed to fill B1 (and I don't), I'd put them in a tax-managed idx stock fund in B3. Ain't puttin' unsheltered money in my bond bucket.

If the stock fund does very well in a year (beating your expected returns) you have some flexibility. For example, I would take half the excess and sweep it into B2 or even B1 if that were running low, to buy some time. Expected 8%, earned 12%, put 12 - 8 = 4% excess, but half that in B2 and let the rest of the excess sit in B3. Or just ignore it if B1 and B2 are still pretty flush - it will only earn you more down the road. It's your call, knowing that there will be down years, too. If I can I like to leave B3 alone. Maybe just cheat by sipping your earnings to take a nice vacation, etc.

[4th&Goal]

Rich,
What percentage of your portfolio is in Bucket #1?

[bobbyrx]

Ray Lucia does say somewhere in his book that you never actually burn through bucket #2 but keep replenishing bucket #1 with rebalancing from equity gains in bucket #3 (assuming that happens). I surmise then that bucket #2 becomes a sort of permanent buffer. This way of thinking suits my personal risk averse nature and has been working for me.

[ddb]

The bucket system seems unnecessarily complicated to me; kind of a "mental accounting" trap. When you combine all three buckets as you lay out, you have a certain aggregate asset allocation, and that's the thing that really matters (not how each bucket is separately allocated).

Having said that, instead of using bucket 1, why not purchase a single-premium immediate annuity with a 10-year guaranteed payout rate (not a lifetime payout). You can buy an income of $1200 per month (no inflation adjustments) for around $122,000. You have now removed all uncertaintly in bucket 1 and saved yourself a few thousand dollars, at the cost of potential upside. Seems pretty reasonable. Of course, the inherent IRR in such a product is only around 3-4%, so you may be uncomfortable locking yourself in to such a low return over ten years.

The above is not a recommendation to implement any particular product or strategy.

- DDB

[Rodc]

Bucket 2 should have nothing more volatile than something along the lines of a 60:40 balanced fund, so I would take the total stock market index and the FTSE EFT out of bucket 2. I like intermediate bonds, TIPs, and Wellesley here, by the way.

I don't really understand the difference between having a 60:40 balanced fund in bucket 2 and having the same amount of stocks and bonds held in two funds, which says TSM and FTSE are fine as long as you include a sufficient amount of bonds.

I agree with ddb that this sounds like mental accounting/implementation crutch to me. If I did the math correctly this starts out about 65% bonds and 35% stocks. You can do the same thing with a simple single account with something like 20% TSM, 15 FTSE, 22% short bonds, 21% intermediate bonds, 22% TIPS, and draw preferentially from short bonds first, etc. In fact a cd/bond/TIPS ladder seems like a good fit here instead of bond funds.

However, I understand there are real benefits to implementation crutches, in fact I might just rely on one when my time comes, so I don't think this is a bad idea.

I fully agree with the posts that one ought to at least consider some type of annuity in place of some of the cash/short bond money.

Best of luck.

[4th&Goal]

Having said that, instead of using bucket 1, why not purchase a single-premium immediate annuity with a 10-year guaranteed payout rate (not a lifetime payout).
- DDB

A single-premium immediate annuity (fixed period) is a Lucia recommendation for Bucket 1.

[CyberBob]

The bucket system seems unnecessarily complicated to me; kind of a "mental accounting" trap. When you combine all three buckets as you lay out, you have a certain aggregate asset allocation, and that's the thing that really matters (not how each bucket is separately allocated).

Lucia's three buckets does seem a but cumbersome when you see similar investments in both bucket 2 and bucket 3.

Personally, I prefer the thinking of Sengupta who works with just 2 buckets. Essentially, his 2 'buckets' are risky assets (i.e. stocks for long term needs) and 'riskless' (or less risky) assets (i.e. money market funds, bonds, TIPS).
He allows for gradations in his 'safe' bucket. So if you have, say, 10 years of withdrawals in that bucket, it would likely have money market funds, shorter-term bond funds, and some TIPS.
He doesn't call them buckets, but rather just views assets by time of need, meaning the money market funds and shorter-term bonds and such are for money you will need soon, and the stocks are for potential greater growth for money that you will not need soon.

Grangaard seems to pretty much thinks in a 2 bucket way as well. (although, the biggest thing he seems to miss is having an actual stock/bond allocation based on something like returns or years of need as Sengupta does).

Bob

[SteveB3005]

ddb writes,
instead of using bucket 1, why not purchase a single-premium immediate annuity with a 10-year guaranteed payout rate (not a lifetime payout). You can buy an income of $1200 per month (no inflation adjustments) for around $122,000.

This struck me as what could be an excellent tool for some people who have not saved a sizable sum for retirement. This could allow those people to delay taking Social Security till 70 years old and collect the substantially higher payout. The people who had not saved a lot would be more than likely working at least part time anyway and the $1200 a month would be a guaranteed monthly boost. What it would do is take the stress out of a possible market crash at that most inopportune time of right before retirement that would cripple what little savings they had.

[williamg]

I don't really understand the difference between having a 60:40 balanced fund in bucket 2 and having the same amount of stocks and bonds held in two funds, which says TSM and FTSE are fine as long as you include a sufficient amount of bonds.

I think the difference lies in the fact that you are minimally rebalancing. A balanced fund will give you a muted return somewhere between all stock or all bond by a process of ongoing rebalancing. To achieve this with combo of stock and bond in B2 you would have to regularly rebalance within that bucket.

[retired at 48]

To: Hudson and Rich in Tampa

My reaction was similar to ddb and Rodc ...why not do same thing in one portfolio...but bucket concepts help clarify what one is doing.

I implemented a similar thing, having to get from age 48 (retired) to age 60 when a pension begins, and age 62 when ssi kicks in. I withdrew $48,000 per year from an IRA using an "annuitized withdrawal formula" , which per IRS regulations avoided the 10% early withdrawal penalty.

I essentially used the "bucket approach". and it worked.

Want more details, see my posting titled "First time poster's IRA at $1,000,000: shares wisdom learned."

Even with a planned downturn in my IRA, it went up!

Retired at 48

[JDeeds]

The bucket system seems unnecessarily complicated to me; kind of a "mental accounting" trap. When you combine all three buckets as you lay out, you have a certain aggregate asset allocation, and that's the thing that really matters (not how each bucket is separately allocated).

Lucia's three buckets does seem a but cumbersome when you see similar investments in both bucket 2 and bucket 3.

Personally, I prefer the thinking of Sengupta who works with just 2 buckets. Essentially, his 2 'buckets' are risky assets (i.e. stocks for long term needs) and 'riskless' (or less risky) assets (i.e. money market funds, bonds, TIPS).
He allows for gradations in his 'safe' bucket. So if you have, say, 10 years of withdrawals in that bucket, it would likely have money market funds, shorter-term bond funds, and some TIPS.
He doesn't call them buckets, but rather just views assets by time of need, meaning the money market funds and shorter-term bonds and such are for money you will need soon, and the stocks are for potential greater growth for money that you will not need soon.

Grangaard seems to pretty much thinks in a 2 bucket way as well. (although, the biggest thing he seems to miss is having an actual stock/bond allocation based on something like returns or years of need as Sengupta does).

Bob

I have read Lucia's books and while his strategy has it's merits, I was completely turned off by his style and investment recommendations. Far too much like a salesperson than a serious planner.

I have more or less independently come around to the strategy you describe as Sengupta's, dividing my assets into Safe and Risky classes. Being the conservative soul that I am, I chose to keep 12 years of anticipated income in my Safe class. This works out to about a 50/50 AA.

I'm funding that with ST and IT government bonds. I estimated the amount to fund that bucket by determining the estimated inflation-adjusted withdrawals and then finding the current cost of buying govt zeros to completely fund it. I also checked the cost of a SPIA and it was lower, but haven't decided to do that yet for any portion of the fund, since I won't formally retire until next year.

Giving myself 12 years guaranteed income allows me to be more aggressive in my stock portfolio. I'm using a globally diversified stock portfolio and it is managed separately - this was inspired by Frank Armstrong's article.

Yes - it's mental accounting and a crutch and all that, but it helps me. Some self-imposed rules are that FI will never fall below 30% of the total portfolio and there will always be at least 6 years of income in the withdrawal account.

[ken250]

hudson,

I think the buckets with stocks in them are relying too heavily on growth.

I would adopt a value/dividend tilt on top of those low income broad stock indexes to supplement the bonds, REITs, cash, etc.

[4th&Goal]

hudson,

I think the buckets with stocks in them are relying too heavily on growth.

I see nothing in Lucia's writing limiting the use of dividend paying stocks or, suggesting only growth stocks.

[Rich_in_Tampa]

Having said that, instead of using bucket 1, why not purchase a single-premium immediate annuity with a 10-year guaranteed payout rate (not a lifetime payout).

Indeed, that is one of the options he mentions for bucket 1. Like everything else, you just decide if 3.5% IRR is worth the certainty, etc. Also, with an IA you have to fudge the inflation part with supplemental withdrawals, but it's a perfectly legitimate choice for bucket 1.

[nisiprius]

Bucket 3 is a diversified stock portfolio. Because it sits for 14 or 15 years, it has enough time to reliably allow you to start the whole process over by reallocating all buckets at that time.

You lost me there. 14 to 15 years?

Bucket 3 needs to appreciate enough to replenish both buckets 1 and 2... in real (inflation-adjusted) value, right?

In 14 to 15 years? Reliably?

$100,000 invested in the stock market in 1961, including reinvested dividends, but corrected for inflation, would have grown to $100,960 in 1975 (14 years) or $129,500 in 1976 (15 years).

And of course, $100,000 invested in the stock market in 1928, including reinvested dividends, but corrected for inflation, would have "grown" to $91,300 in 14 years, $111,400 in 15 years, $135,000 in 20 years.

[Rich_in_Tampa]

Rich, What percentage of your portfolio is in Bucket #1?

I have 6 years of expenses in bucket 1; that drives the amount you need in buckets 2 and 3. A more aggressive investor might have 5 years, versus a very conservative investor might have 10 years.

The actual amount depends on years and also on your chosen withdrawal rate. If you have 6 years and a withdrawal rate of, say, 4.5%, that equates to somewhat less than 27% for B1. Again, sounds like a lot in cash but remember you are burning it down each year.

[Rich_in_Tampa]

Ray Lucia does say somewhere in his book that you never actually burn through bucket #2 but keep replenishing bucket #1 with rebalancing from equity gains in bucket #3 (assuming that happens). I surmise then that bucket #2 becomes a sort of permanent buffer. This way of thinking suits my personal risk averse nature and has been working for me.

Yeah, he's a little coy about that. After mulling this over a while, it seems to me that you are not expected to rebalance conventionally (i.e. every year or so) but only during unusual opportunities (high-return years) and might routinely delay rebalancing for years, depending. It's not a "no touch" approach, but rather a "light touch" approach.

[Rich_in_Tampa]

...why not do same thing in one [or 2] portfolio...but bucket concepts help clarify what one is doing.

Sure, I think you could do this all without 3 buckets, lots of ways to skin this cat.

Personally the buckets work well for me because it clarifies my thinking: I don't take pro-rata withdrawals from balanced funds, I consume my spending money entirely before touching stocks, I have the reassurance of 12 years of expenses taken care of the day I retire, and I am less tempted to buy and sell with the gyrations of the market.

I'm kind of unsophisticated financially, so this resonated with me. For others, the 2 bucket deal might work every bit as well. Whatever floats your boat.

[bobbyrx]

I too sensed that Lucia tends toward infrequent rebalancing except during stellar equity years when you would replenish bucket #1. I like this whole concept although, of course, it is a variation of mental accounting.

Nisiprius is correct that there have been 15 year periods when stocks went nowhere. But I would rather face such a period with 30-35% of my assets in an equity bucket #3 where I have a chance of eventual recovery than in a unified 60/40 equity/bond portfolio.

Bob

[Rich_in_Tampa]

You lost me there. 14 to 15 years?

Bucket 3 needs to appreciate enough to replenish both buckets 1 and 2... in real (inflation-adjusted) value, right?

In 14 to 15 years? Reliably?

$100,000 invested in the stock market in 1961, including reinvested dividends, but corrected for inflation, would have grown to $100,960 in 1975 (14 years) or $129,500 in 1976 (15 years).

And of course, $100,000 invested in the stock market in 1928, including reinvested dividends, but corrected for inflation, would have "grown" to $91,300 in 14 years, $111,400 in 15 years, $135,000 in 20 years.

Well, that's not about buckets, it's about weathering a terrible and prolonged market downturn under any circumstances. If you feel you want a more conservative approach but still like buckets, you just increase the years in bucket 1 to, say, 10 years. That gets you 20 years of fixed income for expenses, but of course at the expense of having less in equities and presumably less opportunity for growth.

Your AA at the time you retire is whatever you want, with or without buckets. The 6 or 7 year bucket 1 above is just an example (akin to a 55:45 allocation, for example). Hope that clarifies.

[ddb]

Another problem I see with the bucket approach (as presented in this thread's original post) is as follows: if you are making your withdrawals for the first 10 years from a bucket which contains entirely cash and fixed income, then your portfolio-level asset allocation (all three buckets combined) is automatically becoming more aggressive (i.e. equity-oriented) each year, even if you rebalance buckets 2 and 3 back to their initial allocations each year. Makes no sense.

- DDB

[bobbyrx]

I think that if you replenish bucket #1 with outsized gains (hopefully) from bucket #3, you will achieve a degree of rebalancing.
But the whole bucket concept, which I am partial too, leads to me to ask a contrarian question.
We are supposed to be tilted more toward equities in early retirement and less so in old age. But I have always wondered whether the reverse is true. If I begin with a $40k draw from a $1million portfolio of equities and have the misfortune of retiring in 1973-1974 or 2000, my initial $40k will be about 5.7% instead of 4% in year one and my portfolio might enter a death-spiral if I don't retrench. But if I am heavier in bond buckets in early retirement and more in equities by age 80 and a dismal period occurs, I should be able to ride out the limited years left and still leave an estate. Or is this totally crazy?

[gkaplan]

I find the whole bucket approach overly complicated and unnecessary. On the other hand, I may feel differently when I retire.

[JDeeds]

Another problem I see with the bucket approach (as presented in this thread's original post) is as follows: if you are making your withdrawals for the first 10 years from a bucket which contains entirely cash and fixed income, then your portfolio-level asset allocation (all three buckets combined) is automatically becoming more aggressive (i.e. equity-oriented) each year, even if you rebalance buckets 2 and 3 back to their initial allocations each year. Makes no sense.

- DDB

This bothered me, too. That's why I decided not set a floor of 30% for FI.

[Rich_in_Tampa]

Another problem I see with the bucket approach (as presented in this thread's original post) is as follows: if you are making your withdrawals for the first 10 years from a bucket which contains entirely cash and fixed income, then your portfolio-level asset allocation (all three buckets combined) is automatically becoming more aggressive (i.e. equity-oriented) each year, even if you rebalance buckets 2 and 3 back to their initial allocations each year. Makes no sense.

Easy to miss in this long thread, but from an earlier post:

The sequence of withdrawing cash first, then bonds, then stocks as a withdrawal strategy (versus conventional pro-rata withdrawals with rebalancing, or selling whatever is high at that time) has support in academic studies - Spitzer et al, Journal of Financial Planning, June 2007.

You might be surprised at the results. In summary, the long, untouched stock growth while cash and bonds are being consumed trumps the risk of being stock-heavy in the middle years.

Anyhow, there are problems with any strategy including Lucia's. In any plan, you have to stay disciplined but nimble, are vulnerablel to long, bad markets no matter what you do, be willing to cut back, etc. etc. as we all know.

[Rich_in_Tampa]

I find the whole bucket approach overly complicated and unnecessary. On the other hand, I may feel differently when I retire.

This thread is much more complicated than the buckets strategy.

And it's a very good thread.

[CyberBob]

Another problem I see with the bucket approach (as presented in this thread's original post) is as follows: if you are making your withdrawals for the first 10 years from a bucket which contains entirely cash and fixed income, then your portfolio-level asset allocation (all three buckets combined) is automatically becoming more aggressive (i.e. equity-oriented) each year, even if you rebalance buckets 2 and 3 back to their initial allocations each year. Makes no sense.

A bucket approach shouldn't automatically become more aggressive as time goes on. That is, if you have 10 years worth of withdrawals in bonds and the rest in stocks, you don't necessarily wait 9 years and 364 days before you think about converting some stocks over to bonds. (Well, you could do it that way, but obviously you are taking a big risk that the stock market might be down at that time).

You have to have some sort of framework for realistic expected returns of the buckets. As time goes by, if the stocks are meeting or exceeding those returns then you should be opportunistic in selling them and putting more money into a safer bond bucket.

For example, my own stock holdings during the 1990's grossly exceeded my expected returns, so I moved more into bonds to ensure a longer period of safe withdrawals (about 15 years in my case). Waiting until all of my bonds were used up and spent before looking to sell stocks would have been quite stupid.

The point of buckets isn't to have buckets filled in some exact way, but simply to have a certain number of withdrawals in safer, less-volatile investments so that when you need them they are there. (akin to a bond ladder). In some bucket strategies, you would actually spend down the stock portion during up stock years and not the bonds.

Bob
(waiting for those 40-year, 4% plus inflation TIPS to go on sale so that I can simplify my buckets down to just one! )

[ddb]

You might be surprised at the results. In summary, the long, untouched stock growth while cash and bonds are being consumed trumps the risk of being stock-heavy in the middle years.

Not surprising at all. Anybody who has run lots of Monte Carlo simulations has seen that the probability of success (for a long-term withdrawal strategy) typically INCREASES as the proportion of equity increases, even at equity levels of 80% or higher.

Of course, the equity proportion which maximizes the probability of future success must be weighted against the investor's risk tolerance. Many retirees wouldn't be able to stomach the volatility of an 80% stock portfolio, even if they are confident that their terminal wealth will be greater.

So going back to the bucket strategy: if such a strategy actually increases your portfolio-level riskiness over time, then why not just make the whole portfolio risky to begin with, and rebalance back to this desired-level-of-riskiness whenever the allocation drifts too far from the target?

I'm a big fan of having an Investment Policy Statement with a target allocation, and sticking with it through the long run. The bucket approach doesn't jibe with the IPS approach, which makes me uncomfortable.

- DDB

[mikenz]

This is a great thread. We should really ask the administrators for a separate Retirement forum.

I think I'll consider some form of a bucket strategy when I retire - I may have to have a backup plan for DW if something happens to me - like "sell everything and put it all into Wellesley" She'd never make any sense of any of this.

[4th&Goal]

Which retirement books talk about the "bucket" system for retirement spending?

The major bucket-type books would probably be:
Buckets of Money, Lucia
The Grangaard Strategy, Grangaard
The Only Proven Road to Investment Success, Sengupta
Asset Dedication, Huxley

And although not the primary focus of the book, I also like the Liquidity Test section on page 173 of Larry Swedroe's The Only Guide to a Winning Investment Strategy You'll Ever Need

Bob

I suggest reading Bob's above listed books. I did and now understand the concept. I didn't understand much of "bucket" type of distributions until reading about the subject. I find this method keeps me disciplined and helps me stay the course. I long ago learned to not fear what I don't understand. Instead, I try to learn more about the subject.

[bob90245]

I'm a big fan of having an Investment Policy Statement with a target allocation, and sticking with it through the long run. The bucket approach doesn't jibe with the IPS approach, which makes me uncomfortable

Correct. However, the distribution phase is different than the accumulation phase. Sticking with a target allocation during the accumulation phase makes good sense. However, when one is kicking that in reverse during the distribution phase, you run into Sequence of Returns risk. The Buckets Strategy is one attempt at mitigating the sequence of returns risk. Another attempt at mitigating the sequence of returns risk is to purchase an immediate fixed annuity with a portion of one's retirement nest egg.

I've written a few articles about this. You can read one of them at the link below. It also talks about sequence of returns risk.

Immediate Annuities in Retirement

[CyberBob]

So going back to the bucket strategy: if such a strategy actually increases your portfolio-level riskiness over time, then why not just make the whole portfolio risky to begin with...

The idea behind buckets isn't to increase riskiness, but to decrease risk due to the fact that you have date-certain available money that isn't susceptible to big stock market downturns.

There seems to be a common misconception about this kind of strategy that it's a 'spend only cash and bonds' strategy*. In fact, during a normal up market period, the majority of your withdrawal would come from selling stocks. That way you would keep your 'safe' assets cushion intact.

The only time where you would be siphoning withdrawals from only the cash and bond side of the allocation would be during a down market. And that's just what the cash and bonds are there for -- so you don't have to sell stocks during a down market.

As for the portfolio becoming more and more risky (I presume you mean more and more biased towards stocks), don't forget that during a down market, the stocks are likely dropping as fast or even faster than you are selling from the bond side of the house.

An example might be the three really bad market years of 2000-2002. If you started out with a 60/40 portfolio, didn't rebalance and took a 4% withdrawal every year by only selling bonds, you actually wound up at the end with a portfolio that was less risky than when you started; 51/49.

Bob

* The simplified explanation of Lucia's system seems to promote this, but the finer details of his strategy actually emphasize opportunistic selling of stocks, rather than just waiting for a specific time and selling a big lump.

[Rodc]

So going back to the bucket strategy: if such a strategy actually increases your portfolio-level riskiness over time, then why not just make the whole portfolio risky to begin with...

The idea behind buckets isn't to increase riskiness, but to decrease risk due to the fact that you have date-certain available money that isn't susceptible to big stock market downturns.

There seems to be a common misconception about this kind of strategy that it's a 'spend only cash and bonds' strategy*. In fact, during a normal up market period, the majority of your withdrawal would come from selling stocks. That way you would keep your 'safe' assets cushion intact.

The only time where you would be siphoning withdrawals from only the cash and bond side of the allocation would be during a down market. And that's just what the cash and bonds are there for -- so you don't have to sell stocks during a down market.

As for the portfolio becoming more and more risky (I presume you mean more and more biased towards stocks), don't forget that during a down market, the stocks are likely dropping as fast or even faster than you are selling from the bond side of the house.

An example might be the three really bad market years of 2000-2002. If you started out with a 60/40 portfolio, didn't rebalance and took a 4% withdrawal every year by only selling bonds, you actually wound up at the end with a portfolio that was less risky than when you started; 51/49.

Bob

* The simplified explanation of Lucia's system seems to promote this, but the finer details of his strategy actually emphasize opportunistic selling of stocks, rather than just waiting for a specific time and selling a big lump.

Those are very good points, and useful to me. I am left though wondering about the difference then of just starting with a hefty amount of cash/short term bonds, etc and a withdrawal strategy of taking from stock gains/dividends first if they exist, and bonds if they don't.

If this is just one method of making sure that happens, I personally think that is fine, many roads to Dublin and all that. Or is there a great deal more to it? (all roads to Dublin differ somewhat, but I'm curious if the divergence is major).

Thanks.

[ddb]

Correct. However, the distribution phase is different than the accumulation phase. Sticking with a target allocation during the accumulation phase makes good sense. However, when one is kicking that in reverse during the distribution phase, you run into Sequence of Returns risk. The Buckets Strategy is one attempt at mitigating the sequence of returns risk. Another attempt at mitigating the sequence of returns risk is to purchase an immediate fixed annuity with a portion of one's retirement nest egg.

Bob, sequence of returns is a risk in both accumulation and distribution phases. In accumulation, you want the good years to happen towards the end (when you have more money). In distribution, you want the good years to happen towards the beginning (when you have more money).

I fail to see how the bucket approach mitigates the sequence of returns risk as well as (or better than) a suitable asset allocation that is rebalanced periodically. Consider a retiree who is 50/50, and in year 1, stock drop 15%. In order to rebalance, said investor will have to make withdrawals from the fixed income portion anyway. In other words, this strategy also doesn't touch stocks during bad times.

As for immediate annuities, I agree that these can be very useful investments tools.

- DDB

[market timer]

The simplified explanation of Lucia's system seems to promote this, but the finer details of his strategy actually emphasize opportunistic selling of stocks, rather than just waiting for a specific time and selling a big lump.

I like market timing as much as the next guy, but let's call a spade a spade here. This buckets method is just an elaborate market timing scheme.