Is "bucketizing" a "MIRAGE" of safety

[Quantum Sufficit]

I am wondering what people think of/how many utilize a safe bucket to spend from while leaving there stock bucket isolated to grow. If you do, do you also periodically refill the safe bucket when you have outsized gains in the stock bucket? (say > 8% per annum, you "scrape the cream off the top" as Ray Lucia says). PLEASE DON'T MAKE THIS THREAD ABOUT RAY LUCIA. There are those on the Boglehead site that say it is a "trick of mental accounting". I can see how it could mitigate sequence of returns IF IT IS USED WITH VALUE AVERAGING ABOVE. IN years with losses, you do not sell any stocks. Please weigh in below!

 

[rayvt]

It's garbage. It is indeed just an illusion of safety. Just consider what happens to your asset allocation in a down market when you're spending from cash. Your allocation tips heavier and heavier to stocks. When the cash bucket is empty you are 100% stocks.

Value Averaging is also an illusion. It only "works" if you ignore scenarios other than the cherry-picked ones.

 

[pb4uski]

I did have a bucket for a short time but decided that it was an accounting mirage. In lieu of a bucket, I refined my 40% fixed income allocation from 40% bonds to 34% bonds and 6% cash and short term investments (which is 18-24 months of living expenses).

So on buckets I agree with rayvt. I don't agree with rayvt that value averaging is an illusion. It works and provides a smidgen better return over DCA. I guess we'll need to agree to disagree on value averaging.

 

[teejayevans]

It's garbage. It is indeed just an illusion of safety. Just consider what happens to your asset allocation in a down market when you're spending from cash. Your allocation tips heavier and heavier to stocks. When the cash bucket is empty you are 100% stocks.

Value Averaging is also an illusion. It only "works" if you ignore scenarios other than the cherry-picked ones.

But the point is the stocks have a chance to recover, you should have both enough cash and bonds to live on until the market recovers. No better example than 2008, if you had 5 years of cash/bonds, you didn't have to sell and take a loss.
The basic concept of having safe money is good, the stuff RL was pushing was junk.
TJ

 

[GrayHare]

But the point is the stocks have a chance to recover, you should have both enough cash and bonds to live on until the market recovers. No better example than 2008, if you had 5 years of cash/bonds, you didn't have to sell and take a loss.

+1

 

[pb4uski]

But if you have even just a 20% allocation to fixed income and a 4% WR, by your definition you didn't have to sell and take a loss. Very few ER have fixed income allocations below 20% - so where's the beef?

 

[Onward]

There are those on the Boglehead site that say it is a "trick of mental accounting".

I would agree.

The only bucket I believe in is the one I will eventually kick.

 

[rayvt]

But the point is the stocks have a chance to recover, you should have both enough cash and bonds to live on until the market recovers. No better example than 2008, if you had 5 years of cash/bonds, you didn't have to sell and take a loss.

Cash earns essentially zero. Actually, after inflation cash has a negative return.
Five years of cash is going to be a huge drag on the overall portfolio performance.

Gee, doesn't anybody bother to use google to investigate ideas they've heard about?
Are Retirement "Bucket" Strategies An Asset Allocation Mirage? - Kitces | Nerd's Eye View

Value Averaging and the Automated Bias of Performance Measures by Simon Hayley :: SSRN

The flaw with Value Averaging is that it insists that your stock portolio grows by a certain cash amount each month. You add or withdraw money to keep it on track. Consider a $100,000 portfolio which you've decided should increase by $1000/mo --- $500 estimated growth and $500 in new money. Then a bear market hits and your stocks drop 3% in a month. Now you have to add $4000 of new money to keep on target. Next month the stocks drop another 3%. You have to add another $4000.

So your budget is set up for monthly deposits of around $500, but you have several months in a row where you are required to put in almost 10 times that.

Consider a year when the stocks lose 15%. You start out with $100,000 and your target for end-of-year is $112,000,of which your expected total new deposits are $6000. But the $100K instead drops to $85K, so you are $27,000 shy and you have to come up with $21,000 more than expected.

Now scale that up to a portfolio that's grown to $200K. You need to add $54,000 to keep in track.

Value Averaging only works on paper because of cherry-picking the presumed market action.

 

[samclem]

Any performance difference between a bucket approach and annual rebalancing likely depends on whether each year's returns are independent (a fresh coin flip) or if they are at least in part influenced by returns from the previous year (or two, or three). If the stock returns for each year are independent, then the two strategies should produce the same results (if the % in each asset category is the same across both approches). OTOH, if several years of poor stock returns make it more likely that a good year will follow (reversion to the mean, or due to changes in P/E or other measures of valuation), then the buckets strategy ought to win out because after a few years of "down" markets the investor has an increasing portion of investments in stocks.
Right?

 

[siamond]

This bucketing thing makes absolutely no sense to me. First, letting big chunks of money in cash or very low yield is just wasting good investment money.

Next, it makes you do judgment calls on what investment vehicles to use in those buckets. Usually driven by perceptions about stocks vs bonds. But who is to say that bonds aren't down at some point in time in the future while stocks are up (hence it'd be much smarter to sell stock)? Notably when you need the money?

Just reverse-balance. It's simple, easy and efficient. You always sell the asset class which is at its highest point, and you keep all your work capital properly invested...

 

[rayvt]

Y'know, there's a reason that Ray Lucia got into trouble. At the heart, the reason he didn't actually have the data to prove that his Buckets method worked is that it doesn't work. Not on real life data. Neither Lucia's Buckets method or the Value Averaging method.

I mean, how hard is it to download historical data and plug it into a spreadsheet?

 

[rgarling]

... didn't actually have the data to prove that his Buckets method worked is that it doesn't work. Not on real life data. ...

I see. Any situation where these systems show a positive return is cherry picked data! By extension, any situation where they show a loss is real life data! Can't argue with that.

 

[pb4uski]

I see. Any situation where these systems show a positive return is cherry picked data! By extension, any situation where they show a loss is real life data! Can't argue with that.

+1

 

[rayvt]

It's easy enough to put together a simple spreadsheet with actual historical data (like the S&P500), and see what it shows.

Although, you don't really need to even go that far. Just explain how Value Averaging works in a bear market when your $200,000 portfolio gets hit for a 20% loss, and you have to add $40,000+ to get back on target. Where does that money come from?

And for Buckets, consider how having 5 years of living expenses in cash is a huge anchor on your portfolio growth.
$4000/mo * 12 * 5yrs = $240,000.
Or, if you are more frugal than me, perhaps $2000/mo, so only $120,000 in cash.

How big is your total portfolio? Having ~25% in non-earning cash is going to kill the performance.

 

[teejayevans]

This bucketing thing makes absolutely no sense to me. First, letting big chunks of money in cash or very low yield is just wasting good investment money.

Next, it makes you do judgment calls on what investment vehicles to use in those buckets. Usually driven by perceptions about stocks vs bonds. But who is to say that bonds aren't down at some point in time in the future while stocks are up (hence it'd be much smarter to sell stock)? Notably when you need the money?

Just reverse-balance. It's simple, easy and efficient. You always sell the asset class which is at its highest point, and you keep all your work capital properly invested...

When you retire you start withdrawing money, so its not about making money, its about preserving it.
If stocks are up, then yes, you sell the stocks, the cash bucket just gives you an option of not selling an investment when it's down, it's more of an insurance policy, and have a couple of years worth of cash (actually CDs, short term bonds, not really cash, investments that tend to maintain value) that get lower returns is going to have minimal impact.
I actually S&P, Bond rate data for the last 100 years into a spreadsheet and played around with having cash reserve vs none and found that it made a significant improvement in performance if retirement started at the beginning of a long downturn (1929, 1966, 1973, 2000).
TJ

 

[rgarling]

Both systems work fine, but are not about optimizing returns, and may not be what you're looking for. Personally, I don't use either one; however, I did make an attempt at value averaging once, but found that I don't have the level of discipline required to execute the system.

Anyhow, as far as value averaging goes, if you had a $240,000 portfolio and experience a 20% loss of the stock portion during one of your accounting periods; you *would* be faced with transferring a little more than 40k to maintain your target return. Many (most?) people who value average contribute to the fixed portion regularly, so that it would tend to grow based on contributions. If so, the money would be available in this example. In essence, with value averaging you are varying your risk profile to maintain a predetermined rate of return (which, of course, might not be realized). To me this example looks a lot like 20% equities, 80% stock with an 'odd' rebalancing technique, which could conceivably end up 100% equities for a time.

Not much to add on buckets of money; however, if it helps someone internalize their level of risk, so much the better. i.e. If their 'safe' bucket will last for 5 years, it might increase their comfort level with their stock market investments.

 

[pb4uski]

It's easy enough to put together a simple spreadsheet with actual historical data (like the S&P500), and see what it shows.

Although, you don't really need to even go that far. Just explain how Value Averaging works in a bear market when your $200,000 portfolio gets hit for a 20% loss, and you have to add $40,000+ to get back on target. Where does that money come from?

And for Buckets, consider how having 5 years of living expenses in cash is a huge anchor on your portfolio growth.
$4000/mo * 12 * 5yrs = $240,000.
Or, if you are more frugal than me, perhaps $2000/mo, so only $120,000 in cash.

How big is your total portfolio? Having ~25% in non-earning cash is going to kill the performance.

The value averaging that I have in mind is just a modified version of dollar cost averaging. So rather than invest $500 at the end of each month, you would invest the amount needed to increase your account balance by $500 from what it was at the beginning of the month (so $500 +/- the change in value for the month). I did this with my taxable accounts and it worked out quite well.

The important point to note is that the amount I invested was the same as DCA, it is just the timing differed and some months I invested more or less than what I would have invested under DCA. In a prolonged bear market cash flow constraints would ultimately cause you to revert to DCA. Similarly, in a prolonged bull market I found I needed to increase my targeted monthly increase to avoid accumulating too much cash and be more fully invested.

As I mentioned earlier, studies have shown that value averaging is slightly better than DCA. See this study.

I agree with you guys that having a target of 5 years in a cash bucket is suboptimal - I target 18-24 months of spending - about half in an online savings account and half in short-term bonds.

 

[siamond]

When you retire you start withdrawing money, so its not about making money, its about preserving it.
If stocks are up, then yes, you sell the stocks, the cash bucket just gives you an option of not selling an investment when it's down, it's more of an insurance policy, and have a couple of years worth of cash (actually CDs, short term bonds, not really cash, investments that tend to maintain value) that get lower returns is going to have minimal impact.
I actually S&P, Bond rate data for the last 100 years into a spreadsheet and played around with having cash reserve vs none and found that it made a significant improvement in performance if retirement started at the beginning of a long downturn (1929, 1966, 1973, 2000).
TJ

Well, sounds to me that you did a bit of data cherry-picking here...

Look, if your premise is indeed strongly oriented towards preservation, then make your primary Asset Allocation very conservative, and you'll get a similar outcome. No need for a bucket approach if that is your goal.

Now if you do keep proper diversity in your AA while doing so (say break down stock in domestic/value/international/emerging in addition to a solid % of bonds), you will actually protect yourself much more effectively. Again, who is to say that the bond market will not go down the drain for whatever reason.

Plus... it also doesn't hurt to keep some potential growth in your portfolio... Could make for rosier retirement years... Or be a buffer against future crisis...

 

[teejayevans]

Well, sounds to me that you did a bit of data cherry-picking here...

Look, if your premise is indeed strongly oriented towards preservation, then make your primary Asset Allocation very conservative, and you'll get a similar outcome. No need for a bucket approach if that is your goal.

Now if you do keep proper diversity in your AA while doing so (say break down stock in domestic/value/international/emerging in addition to a solid % of bonds), you will actually protect yourself much more effectively. Again, who is to say that the bond market will not go down the drain for whatever reason.

Plus... it also doesn't hurt to keep some potential growth in your portfolio... Could make for rosier retirement years... Or be a buffer against future crisis...

1. I assume worst, hope for the best....it's a philosophy that has served me well.
2. We're in a global market, in 2008, everything except gold and treasuries went down, yea I'm diversified in both stocks and bonds, and my cash bucket is really part of my bond allocation, it's not really cash as in $CASH$.
3.I have lots of potential growth, but I invest, so I have/need a long term time horizon, cash buffer gives that.
4.I sleep better knowing I can handle the worst of times, especially with no pension and a global economy that just limps along.
TJ

 

[FANOFJESUS]

I don't know if bucketizing works better but I think it makes some people sleep a lot better at night. In a shorter bear it might work ok. Some of the bad times can last a lot longer than 5 to 7 years.
There were some mean times from 1966 to 1982 I know the stocks were flat and bonds went down. This is a 16 year time frame how would the bucket be holding up now?

 

[youbet]

my cash bucket is really part of my bond allocation, it's not really cash as in $CASH$.

I agree. It's not necessary to have a "cash bucket." Like you, I shun the "cash bucket" concept and instead have adequate stabile and liquid investments inside of my normal AA.

As you say, no need for a "cash bucket."

 

[AndrewJackson]

The bogleheads are right. This is mental accounting bias where an investor divides their money up into groups and thinks each one can accomplish a specified function (safety/growth/high inflation/low inflation/etc.). Instead, you should look at the entire portfolio and allocate it such that the sum of all your investments is in line with your risk tolerance and goals.

 

[rgarling]

The bogleheads are right. ...

The more vocal bogleheads are into theoretical optimums. What if what their 'right' doesn't *feel* right to the individual investor? Do you begrudge him his short-term bucket because he isn't meeting the theoretical definition of optimum? Frankly, this is a case where close counts, so if the investor is using diversified, low cost investment tools, he can't do much better (except by luck).

 

[nun]

I don't see a cash bucket as a hedge against a prolonged down turn as it will eventually run out, income sources that are decoupled from the market are what you want. However, a cash account is convenient and useful for everyday spending and to cover emergencies.

I've always kept my spending and emergency funds in my checking and savings account at my bank. That's my cash bucket and I deal with the negative real return out of habit and convenience. There was a time when I had 6 months of spending in there and it was my entire "portfolio". But, as I saved and the 401k etc grew it became a pretty small percentage of my assets even though it's now around $30k and could support two years of spending as I also get $15k in annual rental income.

My current AA is 50/47/3 and that 3% cash bucket is a buffer for spending and the practicalities of life rather than a hedge against a falling market. For that I have a portion of my fixed income/bonds in a stable value fund, rental income, and eventually SS checks, a small pension and a TIAA Traditional annuity.

 

[mathjak107]

it isn't so much bucketizing that works as it is the fact you are pulling spending cash early on from other places other than your stocks.

if you read the latest study by michael kitces and dr pfau jointly they found anything that lets your stocks grow longer without being liquidated increase your allocation to stocks as the stocks grow and the other money becomes less as it is spent down.

that is pretty conclusive evidence in that study that where you pull from does effect your bottom line differently. if that is what your bucket system does then yes it works.

the bucket system like ray lucias work very differently than the buffer zone study which is very different.

http://www.early-retirement.org/forums/f28/what-makes-annuities-work-more-stocks-67654.html