Page 2 of 6

Posted: Sun May 01, 2005 3:40 pm
by unclemick
Over thirty years ago, I had an epiphany, don't try to convince fellow family members that they were loony tune wack jobs for even listening to Republicans - let alone voting for one. No more politics - great family relationships ever since.

I like JWR's idea of a newsletter format - great place to read and cogitate - or ignore if that's what floats your boat.

If the numbers are relatively meaningless to you - ignore them.

Perhaps a reread of Ben Graham's Intelligent Investor would help.

Who is Reed and who is Kiosaki? And why should I care.

Posted: Sun May 01, 2005 4:40 pm
by JWR1945
It is true that I no longer respond to Norbert Schlenker's posts.

John Walter Russell

Posted: Sun May 01, 2005 4:43 pm
by hix9
hey unclemick / nfs/ ataloss / jwr /es (or anyone else)

still curious as to whether you use(d) lots/little/any tax loss harvesting in your portfolio execution.

i admit up front i don't know anything about it at all so fire away at will.

my impression is a portfolio of higher div. paying stocks, with a goal of living off the dividends, will be exposed to taxes at all times. outside of applying realized capital losses against ordinary income (dividends) to tune of 3,000/yr, is there any way to hide the div's from taxes?

particularly if using reits to 'live off the higher yield' since reit dividends do not qualify for the lower tax rate. now that value funds are running 30% cash i suspect interest on that $ also does not qualify.

recent reduction in tax on dividends mitigates the importance of this for now... maybe not after 2008 though.

a diversified portfolio of low div paying stocks or etf's could seemingly allow one to sell winners and losers simultaneously and avoid taxes (entirely?).

this would require low correlations of course and paper losses would translate into real losses more rapidly- not sure how important this is if the alternative is to 'live off the dividends' of higher yielding stocks since downturns in those stock prices do not create buying opportunities as dividends are used up.

i suspect transaction costs must be higher as well (although discount brokers are cheap nowadays).

i know i am viewing this too simplistically but i am more concerned about tax implications for the future than 'getting the number right' so a good comment or reference would help.

hix9

Posted: Sun May 01, 2005 4:49 pm
by hix9
hey unclemick

Who is Reed and who is Kiosaki? And why should I care.


kiosaki is a writer of popular financial books who had a rich dad and a poor dad.

reed is a real estate investor iirc who slammed much of the contents of kiosaki's writing.

not sure if you should care or not. i think reed felt he was doing a public service by calling attention to some questionable claims kiosaki put forth but since reeds' stuff is free i am a bit biased (never read kiosaki).

hix9

Posted: Sun May 01, 2005 5:08 pm
by JWR1945
ES
I was messing around with this today and my main problem seems to be tying this all in with the ups and downs possible in the market. I don't think I have anything against a straight forward TIPS ladder as part of one's retirement portfolio. But when we go beyond that things get fuzzy.

This is the heart of the problem. Computers don't allow me to get away with hand waving and doing things that are fuzzy.

We know, kind of, what is involved until we start looking really close. When I try to put an algorithm into a calculator, it becomes a lot more difficult.

Please blurt out any ideas that you may have.

I originally thought in terms of P/E10 thresholds for selling principal from the TIPS ladder. But that is not what most people would look at.

Most people would look at last year's gains or losses. I have been looking at an algorithm that sells part of the ladder whenever dividends plus interest fall below a threshold. (The threshold is adjusted to match inflation.)

At this time, I am only removing money from the TIPS ladder, never replenishing it.

I am getting an improvement, but not anywhere nearly as big as I would like. It is of the order of 0.3% [from 3.9% to 4.2%], which is much less than the 1.0%+ that we can get from varying allocations (switching).

The one story that is coming through strongly is that we need to be able to handle the full duration of any dry spell. My data so far favor 20 year ladders or longer (as opposed to 10 or 15 years). IIRC, the US market has had a bad period of 17 years.

It is possible that my calculator isn't doing what I think that it is doing. I have not checked it out to my satisfaction.

One thing that comes to mind is that the nature of dividends has changed in recent decades. If I look at dividends plus interest back in the 1930s and 1940s versus the 1950s versus the 1970s and 1980s, there may be two or three different stories. In the old days, dividend yields were supposed to be high. Until very recently, dividend yields have been allowed to drop quite a bit. Using a single threshold for dividends plus interest may be a mistake.

I am likely to revisit using P/E10 thresholds. They did a great job with switching. Perhaps, they can do better than 0.3% with a TIPS ladder.

Have fun.

John Walter Russell

Posted: Sun May 01, 2005 5:20 pm
by ataloss
hi hix, you are right about the relative tax disadvantages of dividends
although dividends are no longer taxed as ordinary income there is no deferral
no reason not to have some noncorrelated stocks to sell for tax losses to offset gains, some reit distributuions are part dividend and part return of capital, dividend approaches are more appealing in/near the distribution phase (ie with low or no earned income)

unclemick, sorry if you think people are unduly harsh wrt jwr but he can't really explain or defend his "research" and other than a general principles- dividend stocks are good for retirees who need income- you can't really make much of it

you can search for meaning in all this jwr stuff and come up with a little bit of wisdom if you look hard :wink:

anyway h wanted some comments

Posted: Sun May 01, 2005 5:46 pm
by hix9
hi ataloss

i might be dead wrong or maybe i was just unclear.

to clarify, most dividends are not longer taxed at ordinary income rates (reits, other exceptions) but they are also not deductible 1 for 1 against realized capital losses either, correct?

like i said i am clueless here. luckily (rather, unfortunately) won't be an issue for me for years.

i think tax issues might be the one thing to push me towards individual stocks but i think i might overestimate the practical importance for fi/re'ees anyway.

hix9

Posted: Sun May 01, 2005 5:53 pm
by Norbert Schlenker
JWR1945 wrote: It is true that I no longer respond to Norbert Schlenker's posts.

John Walter Russell

Most children discover early that putting fingers in their ears and chanting nanananana doesn't actually change the world. Not our John Walter.

The question is there for all to see, John Walter. It doesn't go away just because you won't answer it. The longer you won't answer, the sorrier you look. Readers are going to think not that you won't but that you can't. In a forum where reputation and expertise depends on answering questions, refusing to answer an on-topic query that uses your own stated assumptions is doom itself.

You're hooped now on the topic of TIPS in an otherwise equity-heavy portfolio. If you post so much as a word about it in any place that allows a response, somebody (not me, I'm done) will again point out the simplest arithmetic consequences of your assumptions and ask a similar question. It will be devastatingly simple and again you will embarrass yourself by being unable to respond.

Poor defenseless John Walter. There he stood, a legend in his own mind, garbed in the finest numeric raiment, until a little boy standing at the edge of the crowd said, "But, Dad, why is he walking around naked?"

Posted: Sun May 01, 2005 6:11 pm
by Norbert Schlenker
hix9 wrote: still curious as to whether you use(d) lots/little/any tax loss harvesting in your portfolio execution.

All the time.
my impression is a portfolio of higher div. paying stocks, with a goal of living off the dividends, will be exposed to taxes at all times. outside of applying realized capital losses against ordinary income (dividends) to tune of 3,000/yr, is there any way to hide the div's from taxes?

No. If you need income to live on and you're not willing to use muni bonds to generate that income, then you're going to pay taxes. Unless you have very high spending, taxes just aren't that high these days. A married couple filing joint can collect $74k in income, take two exemptions and a standard deduction, and still be in the 15% federal bracket. That doesn't depend on the temporarily low dividend tax; that's where federal brackets are on ordinary income.

You may have a different perspective than I do re income levels and tax rates. I think paying $8k in federal tax on $74k in income is a screaming bargain and would not worry one minute about taxes.

Posted: Sun May 01, 2005 6:21 pm
by hix9
hi jwr
Please blurt out any ideas that you may have.


i don't have any ideas but a very naive and poorly thought out question so bear with me.

you propose (if i understand correctly) selling stocks in good (stock market) years and extracting tips principle in bad (stock market). your basis for this is that it would allow you to avoid losing capital in bad (stock market years). this approach seems to maximize tax exposure.

burns wrote (if i understand correctly) to just build an i bond ladder sufficient to cannibalize for the first 10 years of retirement. i think that this approach will minimize taxes as principle is non-taxed while stock portion grows untouched.

why not just take his approach from the beginning?

build a tips ladder such that cannabalizing its interest and principle from day one will support your desired 'rate' (to avoid terminology games) for x years (however many you think you need to weather the worst storm- you say 20) and just start chewing through principle from day 1? as if the worst market for equities 'happened' on the day you retired.

leave equity portion alone (diverse, stocks and etf's) except to sell every losing position (within reason) to harvest tax losses and use proceeds to replinish tips ladder for year 21 etc...

this would allow you to aquire some future tax loss harvesting and keep your 'winner' stocks and etf's tax deferring for 20 years which i think has significant benefits (deferral alpha or something?). if quantifiable might be advantageous.

hix9

Posted: Sun May 01, 2005 6:27 pm
by hix9
hi nfs
You may have a different perspective than I do re income levels and tax rates. I think paying $8k in federal tax on $74k in income is a screaming bargain and would not worry one minute about taxes.


i have similar perspective as you in your example but since i won't retire for decades i doubt it will apply to me.

similarly, i doubt anything i determine today about minimizing (my fear of higher future) taxes will apply then either so i suppose it is an exercise in futility.

then again so was taking latin so i suppose i think even futile ventures can sometimes be educational.

hix9

Posted: Mon May 02, 2005 2:17 am
by ataloss
to clarify, most dividends are not longer taxed at ordinary income rates (reits, other exceptions) but they are also not deductible 1 for 1 against realized capital losses either, correct?


right (and the 3k limitation is ridiculous but we have to live with it)
i think tax issues might be the one thing to push me towards individual stocks but i think i might overestimate the practical importance for fi/re'ees anyway


I ended up with high dividend stocks (like rai) because I was looking for value and I wanted to be able to manage my gains and losses. I think that individual stocks can make a lot of sense although I am keeping my stock funds because of embedded gains. If one were starting fresh (perhaps from winning the lottery) with 1 million to invest in stocks, buying 100 would result in $1000 in comissions once vs essentially the same cost/year in admiral share annual expenses.

nfs, interesting how jwr refuses (ie can't) answer legitimate questions

Posted: Mon May 02, 2005 3:16 am
by ElSupremo
Greetings John :)
Please blurt out any ideas that you may have.

My main problem is time. I don't have any. :( The other problem I ran into was in figuring an adjustment for market volatility. Since there is no way to predict what the market will do that's a tough one. So my first thought was using some kind of average for volatility. But after thinking about that for a while it didn't make any sense since we don't know what Mr. Market is going to do. Catch 22? :( Is it wrong to guess that an extended bear market would wreck the advantages of the TIPS plan?

My third problem is that to have a realistic chance at coming up with any kind of model will take a level of math that I'm not familiar with. So I think it's best to let the math wizards around here have a go at this.

I disagree with those who just write these kind of efforts off. It's easy to trash someone's ideas but much more difficult to pitch in and try to find answers in a constuctive way. I for one appreciate these kinds of efforts in putting forth new ideas. It's how we learn. I wish I had the time and expertise to be of more help.

Thanks John!

Posted: Mon May 02, 2005 3:23 am
by ataloss
so es do you agree with jwr's decision to ignore volatility wrt tips prices?

also are there questions that must not be asked of jwr- might be a help to nfs and myself

won't ask anyone what "insights" they have gleaned from jwr's "research" on this matter :wink:

Posted: Mon May 02, 2005 5:20 am
by unclemick
Duh

I seem to have missed something - ? isn't the whole point of a ladder to sell bonds(of whatever kind) as they come due. Who cares about market vs NAV in the stretch?

Volitility can be friend or foe - depending on your accumulation/distribution cycle and your investment approach. In the distribution phase with 85% trad IRA - I'm bipolar in looking at damping volitility in the IRA to take money out tax efficiently(there are trade offs here) and embracing it for my individual stocks - to get better buys.

To paraphase ancient Ben Graham and Bogle examples - suppose your house value collapsed overnight - would you move out immediately? Versus say your porfolio fell 50% in market value - would you jump for joy at the bargins availible ala Buffett's famous teenager in a cat house quote? Or would you find that mildy upsetting?

As for questions and 'insight' - again back to Ben and Bogle - there's a lot out there that is unknowable. Try to pin down a group of value investors.

To get back to my razz from the 'old forum' - only in context of the total set - up and calculation. My insight - the retirement research is severly constrained by the inflexibility of the math - the burden is on you to apply it to 'your' retirement.

Posted: Mon May 02, 2005 5:54 am
by hix9
hey unclemick
I seem to have missed something - ? isn't the whole point of a ladder to sell bonds(of whatever kind) as they come due. Who cares about market vs NAV in the stretch?


i think that was sort of what i was trying to say a couple posts back- just start out using up the ladder from day1.

i must be completely missing the point of the more complicated approach using market metrics (but that isn't any breaking news).

hix9

Posted: Mon May 02, 2005 6:08 am
by unclemick
The Chinese Box.

Outside of my plunka, plunka DCA, I played with the Chinese box concept I learned from Mr Tuck back in Seattle.

I had a trend line layed out into the future on graph paper (8%) and harvested $ when Mr Market was kind and put it in the credit union(no ladder). Wasted a lot on dirty blondes, some penthouse living and sports cars.

Multi asset and some individual stocks back in those days - never heard of MPT, slice and dice, etc - even though that was what I was doing - including rebalancing before taking money out. Greater or lessor extent thru the 60's, 70's, and 80's.

Box A - could be the ladder.

Box B - could be the first two terms that Bogle uses - with his 'speculative' term the 3rd term swinger(aka Mr Market). Yikes! shades of gummy? Note that B doesn't care whether you slice and dice. balanced index, do real estate, etc. just whether you are above or below trend.

Since the first two terms have an inflation assumption baked in and you have to pick(??6%??) a trend line - I'm not sure what this would tell you.

Anywise - one of things I used to do - before De Gaul and the Norwegian widow aka balanced index and dividend stocks.

Posted: Mon May 02, 2005 6:26 am
by bpp
Sounds kind of like Value Averaging, unclemick. Is that right?

Bpp

Posted: Mon May 02, 2005 6:59 am
by unclemick
No

That would insult value averaging. I was much more eratic - separating adding more to a position, selling and rebalancing, etc, would not fall neatly into a catogory.

Recently per Bernstein's Efficient Frontier website mention, I read Edelson's Value Averaging from the library. And I putzed, putz with Moneypaper's 'Invest%' version with some of my DRIP stocks they follow - again not in a straight forward disciplined mannner - more trying to 'buy value' in addition to reinvested dividends.

Posted: Mon May 02, 2005 10:26 am
by Norbert Schlenker
ES wrote: I disagree with those who just write these kind of efforts off. It's easy to trash someone's ideas but much more difficult to pitch in and try to find answers in a constuctive way. I for one appreciate these kinds of efforts in putting forth new ideas. It's how we learn. I wish I had the time and expertise to be of more help.

Thanks John!

ES, I'm disappointed with this line of thinking.

I quite agree that people should be encouraged to post their ideas for discussion in public places. I quite agree that ideas and effort should not be written off summarily.

However, restricting responses to rah-rah-rah, no matter what idea is presented, is not constructive and does not lead to learning.

Research is not a speech. It's a dialogue in which you should expect to have your assumptions and reasoning applauded by some and questioned by others. While "trashing someone's ideas" is one way of criticizing an argument, it is not the only way.

JWR is of course entitled to make assumptions, do myriad calculations based on those assumptions, and present the results as useful. As a researcher, he then bears the burden of defending the assumptions and the methods he used to create the results. If it then turns out that, using identical assumptions and a much simpler method, one can reach a related but obviously nonsensical conclusion, that indicates a problem.

If you want to learn something, whether you are the researcher or just part of the audience, you must take such criticism to heart. Dismissing it or ignoring it on the grounds that it "trashes someone's idea" is not constructive and makes learning impossible.

There is an elementary principle of logic that, if reasoning properly from a set of assumptions leads to nonsense results, something is wrong with the assumptions. That's all I did in my first post in the thread. I didn't trash JWR's ideas. I just pointed out that the same assumptions, together with a much simpler argument, lead inexorably to nonsense. There is something readers can learn from that, which is that JWR's assumptions must have been nonsense in the first place.

There is also an elementary principle of logic that any conclusion, be it true or false, can be drawn from false assumptions.* There is something that readers can learn from that: Since JWR's assumptions in the thread starter were nonsense to begin with, there is no way to determine whether his results are true or false. It's not that they are necessarily wrong; the problem is that they are useless.

It's your board. You can run it as you like. If you think that it is constructive for someone to repeatedly post useless information, it's your prerogative. I cannot agree. If site policy is going to be

1. Nonsense is permitted, perhaps even encouraged. Repetition especially welcome.
2. Rebutting nonsense is not constructive and will be discouraged.

then the forum is useless. No one can learn anything in such a place.

-----------------------------------------------------------------------

* For those with no exposure to formal logic, here's a simple FIRE/SWR example to demonstrate how proceeding from false assumptions leads to false results. What JWR posts differs only in quantity, not quality.

Assumption: 1% = 0%.

Issue: How much can I withdraw annually from my portfolio without going broke?

Answer: As much as you like. Why? Well, you won't ever go broke if you withdraw 0% from the portfolio. And since 1% = 0%, you won't go broke if you withdraw 1%. And since 2% = 1% + 1% = 0% + 0%, you won't go broke if you withdraw 2%. ... And since 100% = 100 * 1% = 100 * 0% = 0%, you can spend it all in the first year and yet will never go broke!